To begin searching the abstracts, please use the search feature above.
| 1878 | Computer 61
|
Deep Learning Assisted Diagnosis of Prostate Cancer: Using a Multi-scale Neural Network Based on Points of Interest |
| Weiting Huang1, GuoRui Hou 2, Chen Wang3, and Kai Ai4 | ||
1Department of Magnetic Resonance, Lanzhou University Second Hospital, Lanzhou, China, 2Department of Magnetic Resonance, Xijing Hospital, Xi'an, China, 3Department of Radiology, Xijing Hospital, Xi'an, China, 4Philips Healthcare, Xi’an, China |
||
Keywords: Prostate, Machine Learning/Artificial Intelligence, Spatial Transformer Network, Transfer learning In this study, we propose a method to predict clinically significant state cancer based on MRI points of interest (POI) and classification network with multi-mode and multi-scale. Instead of the traditional method of manual delineation region-of-interest (ROI) to assist prediction, our method utilizes multi-scale input combined with Spatial Transformer Network (STN) to automatically adjust the adjust the scale of interest. This work also explored the possibility of predicting the grade of prostate cancer in a small amount of data using the method of transfer learning. Experiments show that this method has high prediction performance. |
| 1879 | Computer 62
|
Age and gender prediction from minimally processed 3D structural brain MRI through multi-task contrastive learning |
| Vick Lau1,2, Christopher Man1,2, Shi Su1,2, Ye Ding1,2, Jiahao Hu1,2, Junhao Zhang1,2, Yujiao Zhao1,2, Alex T. L. Leong1,2, and Ed X. Wu1,2 | ||
1Laboratory of Biomedical Imaging and Signal Processing, The University of Hong Kong, Hong Kong SAR, China, 2Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong SAR, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Predicting brain age from structural MRI (sMRI) is potentially valuable as the deviation of predicted age from chronological age can be a biomarker for characterising brain health conditions. Currently, extensive pre-processing of sMRI data is required for most deep learning methods. This study presents a multi-task contrastive learning framework for simultaneous brain age prediction and gender classification from minimally processed, noisy 3D T1-weighted images. By including gender classification task and supervised contrastive learning, we demonstrate that leveraging gender information in training and better representation learning can boost age prediction accuracy for both in-domain and out-of-domain datasets. |
| 1880 | Computer 63
|
Deep learning-based prognostic model using non-enhanced cardiac cine MRI in patients with heart failure. |
| Yifeng Gao1, Zhen Zhou1, Bing Zhang2, Saidi Guo2, Kairui Bo1, Shuang Li1, Nan Zhang1, Hui Wang1, Yang Guang3, Heye Zhang2, Tong Liu4, Jianxiu Lian5, and Lei Xu1 | ||
1Department of Radiology, Beijing Anzhen Hospital, Beijing, China, 2School of Biomedical Engineering, Sun Yat-Sen University, Guangzhou, China, 3National Heart and Lung Institute, Imperial College London, London, United Kingdom, 4Department of Cardiology, Beijing Anzhen Hospital, Beijing, China, 5Philips Healthcare, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Heart, Deep Learning; Heart Failure We proposed a multi-source deep-learning model including traditional functional parameters and myocardial strain derived from cardiovascular magnetic resonance, as well as clinical features such as laboratory tests, electrocardiograms, and echocardiography. Meanwhile, we innovatively integrated cardiac motion characteristics through deep learning algorithms and fast neural convolution networks to construct deep learning heart failure prediction model. The results showed that compared with the traditional cox model, the deep learning model had higher efficacy for prognosis evaluation in patients with heart failure and could provide risk stratification in patients with heart failure, which may further guide clinical decision-making. |
| 1881 | Computer 64
|
Total Knee Replacement Prediction using Twin Class Distribution Estimation |
| Chaojie Zhang1, Shengjia Chen1, Haoxu Huang2, Haresh Rengaraj Rajamohan3, Jungkyu Park1, Noah Kasmanoff1, Kyunghyun Cho3, Gregory Chang1, Richard Kijowski1, and Cem M. Deniz1 | ||
1Department of Radiology, New York University Langone Health, New York, NY, United States, 2Courant Institute of Mathematical Sciences, New York University, New York, NY, United States, 3Center for Data Science, New York University, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Our study implemented the self-supervised learning method, Twin Class Distribution Estimation, with unlabeled knee MR images. The self-supervised pretraining improves the downstream analysis in predicting total knee replacement within 9 years using labeled knee MR images. The self-supervised features are shown to be efficient classifiers in TKR prediction. |
| 1882 | Computer 65
|
Semi-Supervised Learning with Spatial Pseudo Labeling for Peritumoral Infiltration Prediction in Glioblastoma using MR Fingerprinting |
| Walter Zhao1, Xiaofeng Wang2, Charit Tippareddy3, Hamed Akbari4,5, Anahita Fathi Kazerooni4,5, Christos Davatzikos4,5, Marta Couce6, Andrew E. Sloan6,7,8,9, Chaitra Badve3, and Dan Ma1 | ||
1Department of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States, 2Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH, United States, 3Department of Radiology, Case Western Reserve University and University Hospitals Cleveland Medical Center, Cleveland, OH, United States, 4Center for Biomedical Image Computing and Analytics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, United States, 5Department of Radiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, United States, 6Department of Pathology, Case Western Reserve University and University Hospitals Cleveland Medical Center, Cleveland, OH, United States, 7Department of Neurosurgery, Case Western Reserve University and University Hospitals Cleveland Medical Center, Cleveland, OH, United States, 8Seidman Cancer Center and Case Comprehensive Cancer Center, Cleveland, OH, United States, 9Piedmont Health, Atlanta, GA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Cancer Existing glioblastoma (GB) infiltration models are often limited by lack of true infiltration labels and employ the assumption that edema closer to tumor has higher infiltrative potential relative to distant edema. Here, we propose a semi-supervised learning scheme that incorporates pretraining on the near-far heuristic and spatial pseudo labeling using true infiltration labels for voxel-wise tumor infiltration prediction. Our results show improved classification performance following finetuning on labeled infiltration data compared to training on the near-far heuristic alone and indicate the potential in employing MR fingerprinting-based models to guide GB diagnosis and treatment. |
| 1883 | Computer 66
|
Biparametric MRI classification model for prostate cancer detection using a combination of prediction maps |
| Mohammed R. S. Sunoqrot1,2, Rebecca Segre1,3, Gabriel A. Nketiah1,2, Alexandros Patsanis1, Tone F. Bathen1,2, and Mattijs Elschot1,2 | ||
1Department of Circulation and Medical Imaging, Norwegian University of Science and Technology (NTNU), Trondheim, Norway, 2Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway, 3Department of Mechanical and Aerospace Engineering, Politecnico di Torino, Torino, Italy |
||
Keywords: Machine Learning/Artificial Intelligence, Prostate Biparametric MRI (bpMRI) is a valuable tool for the diagnosis of prostate cancer (PCa). Computer-aided detection and diagnosis (CAD) systems have the potential to improve the robustness and efficiency of PCa detection) compared with conventional radiological reading. This work explores the combination of the results of two PCa CAD systems as input to a new classification model. We show that this is promising approach that can potentially improve the final detection of PCa. |
| 1884 | Computer 67
|
Prediction of IDH Mutation Status of Gliomas Using Pre-operative MR Images: a Fully Automatic Convolutional Neural Networks Based Approach. |
| Xiaohua Chen1, Zhiqiang Chen2, Zhuo Wang1, Shaoru Zhang1, Yunshu Zhou1, Shili Liu1, Ruodi Zhang1, Yuhui Xiong3, and Aijun Wang4 | ||
1Clinical medicine school of Ningxia Medical University, Yinchuan, China, 2Department of Radiology ,the First Hospital Affiliated to Hainan Medical College, Haikou, China, 3GE Healthcare MR Research, Beijing, China, 4Department of Radiology, General Hospital of Ningxia Medical University, Yinchuan, China |
||
Keywords: Machine Learning/Artificial Intelligence, Brain Using an automated method of convolutional neural networks (CNNs), we aimed to predict the IDH mutation status of gliomas from conventional preoperative MRI in this study. We conclude that the Markov Random Field-U-Net network can accurately segment the tumor region. Using a modified 34-layer Resnet network, we were subsequently able to predict IDH mutation status effectively. Consequently, the model has the potential to be utilized more broadly as a practical tool with reproducibility, this model has the potential to be a practical tool for the non-invasive characterization of gliomas to help the individualized treatment planning. |
| 1885 | Computer 68
|
Prediction of active Multiple Sclerosis lesions through use of logistic regression classifier and first-order features |
| Vivian S. Nguyen1,2, Adam J. Hasse3, Emily Tao1, Jihye Jang4, Adil Javed5, Timothy J. Carroll3, and Keigo Kawaji1,2 | ||
1Biomedical Engineering, Illinois Institute of Technology, Chicago, IL, United States, 2Medicine - Cardiology, University of Chicago Medical Center, Chicago, IL, United States, 3Radiology, University of Chicago Medical Center, Chicago, IL, United States, 4Philips Healthcare, Gainesville, FL, United States, 5Neurology, University of Chicago Medical Center, Chicago, IL, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Multiple Sclerosis Multiple Sclerosis is a neuroinflammatory disease in which the immune system attacks nerve fibers and myelin sheaths, leading to the formation of lesions through white matter. Gadolinium-enhanced MRI is used to diagnose and track the progression of MS. Active MS lesions enhance with gadolinium, but there is an interest in prediction of lesion enhancement based on lesion features. In this study, we examined first-order features derived from T1w pre-contrast MS lesions acquired on multiple 3T imagers at a single center to train a logistic regression classifier to classify lesions as active or inactive. |
| 1886 | Computer 69
|
Prediction of ductal carcinoma in situ with microinvasion postoperatively in women with biopsy-confirmed ductal carcinoma in situ |
| Zhou Huang1, Xue Chen2, Nan Jiang1, Su Hu1, and Chunhong Hu1 | ||
1Department of Radiology, The First Affiliated Hospital of Soochow University, Suzhou, China, 2Department of Radiology, The Affiliated Suzhou Hospital of Nanjing Medical University, Suzhou, China |
||
Keywords: Machine Learning/Artificial Intelligence, Breast To predict ductal carcinoma in situ with microinvasion (DCISMI) based on clinicopathologic, conventional breast magnetic resonance imaging (MRI), and dynamic contrast enhanced MRI (DCE-MRI) radiomics signatures in women with biopsy-confirmed ductal carcinoma in situ (DCIS) to choose high-risk women who may benefit from sentinel lymph node biopsy at initial surgery. The mixed model showed better AUC values than both clinicopathological and DCE-MRI radiomics models in both training/test sets with heterogeneous enhancement and radiomics scores as significant independent predict factors. The mixed model showed the greatest overall net benefit for upstaging and the second was the combine model. |
| 1887 | Computer 70
|
Development and validation of a deep learning model trained on MRI for the prediction of hepatocellular carcinoma survival |
| Lidi Ma1, Congrui Li2, Haixia Li3, Kan Deng3, Cheng Zhang1, Weijing Zhang1, and Chuanmiao Xie1 | ||
1Department of Radiology, Sun Yat-sen University Cancer Center, Guangzhou, China, 2Department of Diagnostic Radiology, Hunan Cancer Hospital, Central South University, Changsha, China, 3Philips Healthcare, Guangzhou, China |
||
Keywords: Machine Learning/Artificial Intelligence, Liver In this study, we developed a deep learning model based on GD-DTPA-enhanced MRI data to predict the overall survival (OS) of patients with HCC. Our results showed that 3D-CNN model based on GD-DTPA-enhanced MRI can non-invasively predict the OS of patients with HCC. The combined model integrating the deep learning score and clinical factors showed a higher predictive value than the clinical and 3D-CNN models and may be more useful in guiding clinical treatment decisions to improve the prognosis of patients with HCC. |
| 1888 | Computer 71
|
Machine learning methods for cerebral perfusion status prediction |
| Linkun Cai1, Haijun Niu1, Erwei Zhao2, Yawen Liu1, Tingting Zhang1, Dong Liu3, Penggang Qiao4, Pengling Ren4, Wei Zheng2, and Zhenchang Wang4 | ||
1School of Biological Science and Medical Engineering,Beihang University, Beijing, China, 2National Space Science Center,Chinese Academy of Sciences, Beijing, China, 3Department of Ultrasound, Beijing Friendship Hospital, Beijing, China, 4Department of Radiology, Beijing Friendship Hospital, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence The diagnosis and evaluation of cerebral perfusion status are crucial for the management of brain diseases. However, the detection method of cerebral perfusion status is complicated. Considering that CBF is mainly supplied by the internal carotid artery (ICA), this paper proposes a novel cerebral perfusion status prediction model, which can automatically quantify the cerebral perfusion level of patients by modeling the association between ICA blood flow and cerebral perfusion. The experimental results on a real-world dataset using machine learning methods can achieve satisfactory performance. Thus, it can be used as an effective adjuvant tool for determining the cerebral perfusion status. |
| 1889 | Computer 72
|
BayesNetCNN: incorporating uncertainty in neural networks for image-based classification tasks |
| Matteo Ferrante1, Tommaso Boccato2, Marianna Inglese2, and Nicola Toschi3,4 | ||
1Biomedicine and prevention, University of Rome Tor Vergata, Roma, Italy, 2Biomedicine and prevention, University of Rome Tor Vergata, Rome, Italy, 3BioMedicine and prevention, University of Rome Tor Vergata, Rome, Italy, 4Department of Radiology,, Athinoula A. Martinos Center for Biomedical Imaging and Harvard Medical school, Boston, MA, USA, Boston, MA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Alzheimer's Disease The willingness to trust predictions formulated by automatic algorithms is key in a vast number of domains. We propose converting a standard neural network into a Bayesian neural network and estimating the variability of predictions by sampling different networks at inference time. We use a rejection-based approach to increase classification accuracy from 0.86 to 0.95 while retaining 75% of the test set for Alzheimer disease classification from MRI morphometry images. Estimating uncertainty of a prediction and modulating the behavior of the network to a desirable degree of confidence, represents a crucial step in the direction of responsible and trustworthy AI. |
| 1890 | Computer 73
|
Unsupervised Domain Adaptation for Automated Knee Osteoarthritis Phenotype Classification |
| Junru Zhong1, Yongcheng Yao1, Dόnal G. Cahill2, Fan Xiao3, Siyue Li1, Jack Lee4, Kevin Ki-Wai Ho5, Michael Tim-Yun Ong5, James F. Griffith2, and Weitian Chen1 | ||
1CU Lab for AI in Radiology (CLAIR), Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong, Sha Tin, NT, Hong Kong, 2Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong, Sha Tin, NT, Hong Kong, 3Department of Radiology, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, Shanghai, China, 4Centre for Clinical Research and Biostatistics, The Chinese University of Hong Kong, Sha Tin, NT, Hong Kong, 5Department of Orthopaedics & Traumatology, The Chinese University of Hong Kong, Sha Tin, NT, Hong Kong |
||
Keywords: Osteoarthritis, Osteoarthritis We propose a knee osteoarthritis phenotype classification system using unsupervised domain adaptation (UDA). A convolutional neural network was initially trained on a large source dataset (Osteoarthritis Initiative, n=3116), then adapted to a small target dataset (n=50). We observed a significant performance improvement compared to the classifiers trained solely on the target dataset. We demonstrated the feasibility of applying UDA for medical image analysis in a small label-free dataset. |
| 1891 | Computer 74
|
RiskForm: A novel risk formulation to improve progressive disease outcome prediction |
| Haresh Rengaraj Rajamohan1, Kyunghyun Cho1, Richard Kijowski2, and Cem M. Deniz2 | ||
1New York University, NEW YORK, NY, United States, 2NYU Langone Health, NEW YORK, NY, United States |
||
Keywords: Osteoarthritis, Machine Learning/Artificial Intelligence, Deep Learning We propose a novel risk constraint to improve the performance of deep learning models on progressive disorders. On the Osteoarthritis Initiative (OAI) dataset, the proposed approach outperforms a baseline model trained with the standard cross-entropy loss on predicting total knee replacement (TKR) within 3 different time horizons- 1 year, 2 years and 4 years of the MRI date. It further generalizes better to the external Multicenter Osteoarthritis Study dataset. |
| 1892
|
Computer 75
|
Neural Shape Models Predict Knee Pain Better than Conventional Statistical Shape Models: Data from the Osteoarthritis Initiative |
| Anthony A Gatti1, Dave Van Veen2, Garry E Gold1, Scott L Delp3, and Akshay S Chaudhari1 | ||
1Radiology, Stanford University, Stanford, CA, United States, 2Electrical Engineering, Stanford University, Stanford, CA, United States, 3Bioengineering, Stanford University, Stanford, CA, United States |
||
Keywords: Osteoarthritis, Machine Learning/Artificial Intelligence MRI-based statistical shape models can predict future disease and distinguish between patient groups. However, these models require thousands of matching points between bones which may introduce biases and their strictly linearly orthogonal features is a limitation. This study built continuous 3D shape representations of the femur using neural implicit representations and used the learned latent space to predict knee pain. The neural shape model can generate arbitrarily high resolution surfaces and predict pain with area under the receiver operating characteristic curve of 0.7 and sensitivity of 0.89, metrics comparable to deep learning methods trained on orders of magnitude more data. |
| 1893 | Computer 76
|
Multi-Modal Detection and Localization of Intracranial Aneurysms using 3D nnDetection Deep Learning Model |
| Maysam Orouskhani1, Shaojun Xia2, Mahmud Mossa-Basha3, and Chengcheng Zhu3 | ||
1University of Washington, SEATTLE, WA, United States, 2Peking University Cancer Hospitals & Institution, Beijing, China, 3Department of Radiology, University of Washington, Seattle, WA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Aneurysm detection, nnDetection, Aneurysm Localization Intracranial aneurysms are relatively common life-threatening diseases with a prevalence of 3.2% in the general population. Therefore, detection is a vital task in aneurysm management. Lesion detection refers to simultaneously localizing and categorizing the lesions in medical images. In this study, we employed nnDetection framework, a self-configuring framework for 3D medical object detection, to detect and localize the 3D coordination of aneurysms. To capture and extract diverse features of aneurysms, two modalities including TOF-MRA, and structural MRI from ADAM dataset have been used. The performance of the proposed deep learning model was evaluated by free-response receiver operative characteristics |
| 1894 | Computer 77
|
Estimating time-to-total knee replacement surgery using deep learning |
| Eisa Hedayati1, Haresh Rajamohan2, Lily Zhou3, Kyunghyun Cho2, Gregory Chang1, Richard Kijowski1, and Cem M Deniz1 | ||
1Radiology, New York University Langone health, New York, NY, United States, 2Center for data Science, New York University, New York, NY, United States, 3Radiology & diagnostic Imaging, University of Alberta, Edmonton, AB, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Osteoarthritis "When do I need my knee replaced? " is a common question of patients with progressed osteoarthritis conditions. However, answering that question is not trivial, especially for cases that do not result in an immediate surgery. To help doctors addressing this question we employed neural networks to recommend an estimated subject-specific date for the total knee replacement (TKR). |
| 1895 | Computer 78
|
GAMER-MRI and modified Layer-wise Relevance Propagation identify on quantitative MRI regions sensitive to clinical disability in MS patients |
| Po-Jui Lu1,2,3, Benjamin Odry4, Muhamed Barakovic1,2,3, Matthias Weigel1,2,3,5, Robin Sandkühler6, Reza Rahmanzadeh7, Xinjie Chen1,2,3, Mario Ocampo-Pineda1,2,3, Jens Kuhle2,3, Ludwig Kappos2,3, Philippe Cattin6, and Cristina Granziera1,2,3 | ||
1Translational Imaging in Neurology (ThINk) Basel, Department of Medicine and Biomedical Engineering, University Hospital Basel and University of Basel, Basel, Switzerland, 2Department of Neurology, University Hospital Basel, Basel, Switzerland, 3Research Center for Clinical Neuroimmunology and Neuroscience Basel (RC2NB), University Hospital Basel and University of Basel, Basel, Switzerland, 4AI for Clinical Analytics, Covera Health, New York, NY, United States, 5Division of Radiological Physics, Department of Radiology, University Hospital Basel, Basel, Switzerland, 6Center for medical Image Analysis & Navigation, Department of Biomedical Engineering, University of Basel, Allschwil, Switzerland, 7Neuroradiology Department, Inselspital, Bern, Switzerland |
||
Keywords: Machine Learning/Artificial Intelligence, Multiple Sclerosis The decision process of artificial intelligence is elusive. We proposed a new method that by combining an attention-based convolutional neural network (GAMER-MRI) with the modified Layer-wise Relevance Propagation could reveal relevant regions on quantitative imaging maps in differentiating multiple sclerosis patients with mild-moderate and severe disabilities. The assessment of the relevant regions included the impact of inverting values within the regions and the heatmap on the MNI152 template. Our results show good network performance and identify brain regions relevant to the corticospinal tract. The proposed method might be useful to further explore patterns of brain microstructural alterations associated with disability. |
| 1896 | Computer 79
|
Pre-Clinical MRI radiomics and machine learning to predict survival after immunotherapy treatment |
| Vlora Riberdy1, Alessandro Guida2,3, James Rioux1,2,3, and Kimberly Brewer1,2,3,4,5 | ||
1Physics and Atmospheric Science, Dalhousie University, Halifax, NS, Canada, 2Diagnostic Radiology, Dalhousie University, Halifax, NS, Canada, 3Biomedical MRI Research Laboratory, Nova Scotia Health Authority, Halifax, NS, Canada, 4School of Biomedical Engineering, Dalhousie University, Halifax, NS, Canada, 5Microbiology & Immunology, Dalhousie University, Halifax, NS, Canada |
||
Keywords: Molecular Imaging, Radiomics Molecular MRI allows for immunotherapy treatment monitoring of glioblastoma, but analysis of multi-parametric data is complex. Machine learning algorithms can be applied to quantitative maps, to identify correlations between radiomic features and treatment outcomes. Feature selection is key when dealing with longitudinal preclinical data with multiple contrasts and small group numbers. We evaluated three feature selection methods in terms of their ability to produce predictive models of survival. The best performance was seen using recursive feature elimination applied to features from iron concentration maps of the tumor, which yielded an ROC AUC of 0.78 and an accuracy of 0.72. |
| 1897 | Computer 80
|
Towards an optimal breast lesions predictive model by assessing different MRI protocols’ combinations: Radiomics analysis |
| Gelareh Valizadeh1, Fereshteh Khodadadi Shoushtari2, Soheila Koopaee1, Hanieh Mobarak Salari1, Mohammad Hossein Golezar3, Masomeh Gity4, and Hamidreza Saligheh Rad1,5 | ||
1Quantitative MR Imaging and Spectroscopy Group, Tehran University of Medical Sciences, Tehran, Iran (Islamic Republic of), 2Shiraz University, Shiraz, Iran (Islamic Republic of), 3Shahed University, Tehran, Iran (Islamic Republic of), 4Tehran university, Tehran, Iran (Islamic Republic of), 5Department of Medical Physics and Biomedical Engineering, Tehran University of Medical Sciences, Tehran, Iran (Islamic Republic of) |
||
Keywords: Multimodal, Breast This study aimed to assess added value of various combinations of different MRI protocols and artificial intelligence techniques in differentiation capability between malignant and benign breast lesions. 61 benign and 69 malignant lesions were recruited. Radiomics features were extracted from three proposed scenarios, including original images of ADC and T2W (scenario-I), original images of ADC, T2W, and DCE (scenario-II), and the joint of original and the pre-filtered images (using wavelet) from ADC, T2W and DCE (scenario-III). Ten most relevant features were utilized for training 11 machine learning algorithms. Finally, decision tree achieved the highest results of accuracy using scenario-III. |
| 3701 | Computer 121
|
Comparison of data-driven and physics-informed learning approaches for optimising multi-contrast MRI acquisition protocols |
| Álvaro Planchuelo-Gómez1,2, Maxime Descoteaux3, Santiago Aja-Fernández2, Jana Hutter4, Derek K. Jones1, and Chantal M.W. Tax1,5 | ||
1CUBRIC, Cardiff University, Cardiff, United Kingdom, 2Imaging Processing Laboratory, Universidad de Valladolid, Valladolid, Spain, 3SCIL, Université de Sherbrooke, Sherbrooke, QC, Canada, 4Centre for Medical Engineering, Centre for the Developing Brain, King's College London, London, United Kingdom, 5Image Sciences Institute, University Medical Centre Utrecht, Utrecht University, Utrecht, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Data Acquisition, Quantitative Imaging Multi-contrast MRI is used to assess the biological properties of tissues, but excessively long times are required to acquire high-quality datasets. To reduce acquisition time, physics-informed Machine Learning approaches were developed to select the optimal subset of measurements, decreasing the number of volumes by approximately 63%, and predict the MRI signal and quantitative maps. These selection methods were compared to a full data-driven and two manual strategies. Synthetic and real 5D-Diffusion-T1-T2* data from five healthy participants were used. Feature selection via a combination of Machine Learning and physics modelling provides accurate estimation of quantitative parameters and prediction of MRI signal. |
| 3702 | Computer 122
|
Predicting Abdominal MRI Protocols using Electronic Health Records |
| Peyman Shokrollahi1, Juan M. Zambrano Chaves1, Avishkar Sharma1, Jonathan P.H. Lam1, Debashish Pal2, Naeim Bahrami2, Akshay S. Chaudhari1, and Andreas M. Loening1 | ||
1Radiology, Stanford University, Stanford, CA, United States, 2GE Healthcare, Sunnyvale, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Modelling, MR Radiology Workflow Inaccurate selection of MRI protocols can impede diagnostics and therapeutic workflows, delay appropriate treatment, increase misdiagnosis likelihoods, and increase healthcare costs. Here we depict a machine-learning (ML) based system to accurately predict abdominal MR protocols, trained on the electronic medical records from 11,251 MR exam orders on 6,882 patients. The best model achieved a cumulative F1-score of 95.6% for top-three most-often-ordered protocols and a top-one F1 score of 78.5%. The proposed system can guide radiologists to appropriate protocol selections quickly, optimize workflows, and improve diagnostic accuracy, thereby by serving to support optimal patient outcomes. |
| 3703 | Computer 123
|
Theoretical guaranteed unfolding method for k-space interpolation in a self-supervised manner |
| Chen Luo1,2, Zhuoxu Cui2, Huayu Wang1,2, Qiyu Jin1, Guoqing Chen1, and Dong Liang2 | ||
1Inner Mongolia University, Hohhot, China, 2Shenzhen Institute of Advanced Technology，Chinese Academy of Sciences, Shenzhen, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Recently, iterative algorithm driven deep neural network unfolding methods have been successfully applied to MRI. However, the network replaces the original algorithm structure and mathematical properties such as interpretability and convergence of the original algorithm are not guaranteed. Fortunately, the k-space-filled Hankel low rank can naturally be associated with convolutional networks. Given this, we propose an unfolding method for k-space-filling, which guarantees convergence to the unique real MR image. Furthermore, we train this network in a self-supervised manner to cope with scenarios where fully sampled data are difficult to obtain. Finally, numerical experiments validate the effectiveness of the proposed method. |
| 3704 | Computer 124
|
Deep learning-based acquisition protocol optimization and parameter estimation for diffusion exchange spectroscopy |
| Zhaowei Cheng1, Guangxu Han2, Songtao Hu2, Ke Fang1, and Ruiliang Bai3 | ||
1College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China, 2Key Laboratory of Biomedical Engineering of Ministry of Education, College of Biomedical Engineering and Instrument Science, Zhejiang University, Hangzhou, China, 3School of Medicine, Zhejiang University, Hangzhou, China |
||
Keywords: Machine Learning/Artificial Intelligence, Data Acquisition We propose a deep-learning-based method to optimize acquisition protocol and estimate parameter for diffusion exchange spectroscopy. A unified framework has been carefully designed to achieve both goals simultaneously. Using this framework, the acquisition protocol can be directly optimized with an objective function that minimizes the parameter estimation errors, regardless of its degree of freedom. The experimental results from Monte Carlo simulations show that the proposed method outperforms the existing methods in both accuracy and precision of parameters’ estimation. Besides, it speeds up the calculation by 480 times. |
| 3705 | Computer 125
|
Feasibility of Deep Learning Reconstruction in Prostate Multiparametric MRI: a Preliminary Prospective Study |
| Yichen Wang1, Xinxin Zhang1, Sicong Wang2, Xinming Zhao1, and Yan Chen1 | ||
1Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy, Beijing, China, 2GE Healthcare, MR Research, Beijing, China |
||
Keywords: Prostate, Machine Learning/Artificial Intelligence, Deep learning reconstruction In this prospective study, feasibility of deep learning reconstruction (DLR) in axial FSE-T2WI and axial reduced-FOV DWI (FOCUS DWI) were evaluated compared with standard protocols. Fast protocol with DLR substantially reduced scanning time (axial FSE-T2WI: -32.1%; FOCUS-DWI: -36.8%). Fast FOCUS DWI with DLR showed the highest SNR and CNR for prostate PZ, TZ and lesion. Fast FSE-T2WI with DLR showed the highest SNR and CNR for prostate PZ and TZ. Moreover, fast FOCUS-DWI and FSE-T2WI with DLR demonstrated equivalent or better image quality than standard images. DLR may be useful in prostate multiparametric MRI protocol optimization and high-quality image acquisition. |
| 3706 | Computer 126
|
Deep Learning Augmented PROPELLER Reconstruction for Improved MRI Motion Correction |
| Sixing Liu1, Lifeng Mei1, Shoujin Huang1, and Mengye Lyu1 | ||
1College of Health Science and Environmental Engineering, Shenzhen Technology University, Shenzhen, China |
||
Keywords: PET/MR, Machine Learning/Artificial Intelligence, deep learning Applying the Periodically Rotated Overlapping Parallel Lines with Enhanced Reconstruction (PROPELLER) technique is one of the strategies to mitigate motion artifacts in MR images. However, due to technical limitations, existing method estimates motion parameters with unsatisfactory results, motion artifacts will still be presented in the final image. Deep learning algorithms are expected to optimize the motion parameter estimation part of PROPELLER technique. We develop a PROPELLER imaging technique incorporating a deep learning model that can provide accurate results and greatly shorten the elapsed time. |
| 3707 | Computer 127
|
Deep learning reconstruction algorithm for 100-second rapid Ischemic stroke imaging |
| Xin Fang1, Ailian Liu1, Qingwei Song1, Guobin Li2, and Shuheng Zhang2 | ||
1the First Affiliated Hospital of Dalian Medical University, Dalian, China, 2Shanghai United Imaging Healthcare Co., Ltd, Shanghai, China |
||
Keywords: Brain Connectivity, Brain Head Computed Tomography (HCT) is a common imaging method for the diagnosis of emergency cerebral infarction. However, it is easy to miss the diagnosis due to the poor performance of the ultra-early or early cerebral infarction lesions. Compared with CT, head Magnetic Resonance Imaging (MRI) has the characteristics of high tissue contrast, no ionizing radiation damage and can make functional imaging sequences, so it has significant advantages in the diagnosis of posterior fossa lesions and ultra-early cerebral infarction |
| 3708 | Computer 128
|
Perturbation loss with carrier image reconstruction: A loss function for optimized point spread functions |
| R. Marc Lebel1,2 | ||
1GE Healthcare, Calgary, AB, Canada, 2Radiology, University of Calgary, Calgary, AB, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Deep learning (DL) assisted image reconstructions are becoming state-of-the art, producing better image quality and/or enabling higher acceleration rates than achievable with conventional methods. DL networks are used to mitigate noise amplification while retaining important signal characteristics. However, typical loss functions produce object-dependent noise alterations and non-uniform point-spread functions. Here we present a method for training networks that prioritizes maximizing the point spread function to ensure maximal detail retention. |
| 3709 | Computer 129
|
U-JET: Preliminary results of a convolutional neural network approach for distortion-free image reconstruction of PROPELLER data |
| Jörn Huber1, Klaus Eickel1,2, and Matthias Günther1,2,3 | ||
1Fraunhofer Institute for Digital Medicine MEVIS, Bremen, Germany, 2mediri GmbH, Heidelberg, Germany, 3Faculty 1 (Physics/Electrical Engineering), University of Bremen, Bremen, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Artifacts Arterial Spin Labeling has great potential in clinics as a non-invasive alternative to Dynamic Contrast Enhanced imaging. However, motion sensitivity needs to be tackled by readout techniques such as 3D GRASE PROPELLER, which unfortunately shows a high sensitivity to geometric distortion. Analytical separation of motion and distortion effects is computationally demanding and might fail in some situations. To this aim, a U-NET based convolutional neural network approach is demonstrated which might overcome this limitation. |
| 3710 | Computer 130
|
A Faithful Deep Sensitivity Estimation Makes High-quality MRI Reconstruction |
| Zi Wang1, Haoming Fang1, Chen Qian1, Boxuan Shi1, Lijun Bao1, Liuhong Zhu2, Jianjun Zhou2, Wenping Wei3, Jianzhong Lin4, Di Guo5, and Xiaobo Qu1 | ||
1Department of Electronic Science, Biomedical Intelligent Cloud R&D Center, Fujian Provincial Key Laboratory of Plasma and Magnetic Resonance, National Institute for Data Science in Health and Medicine, Xiamen University, Xiamen, China, 2Department of Radiology, Zhongshan Hospital (Xiamen), Fudan University, Xiamen, China, 3Department of Radiology, Xiamen University, Xiamen, China, 4Department of Radiology, Zhongshan Hospital affiliated to Xiamen University, Xiamen, China, 5School of Computer and Information Engineering, Xiamen University of Technology, Xiamen, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Recent deep learning is superior in providing high-quality images and ultra-fast reconstructions in accelerated magnetic resonance imaging (MRI). Faithful coil sensitivity estimation is vital for MRI reconstruction. In this work, we propose a Joint Deep Sensitivity estimation and Image reconstruction network (JDSI). During the image artifacts removal, it gradually provides more faithful sensitivity maps, leading to greatly improved image reconstructions. Results on in vivo datasets and radiologist reader study demonstrate that, the proposed JDSI achieves the state-of-the-art performance visually and quantitatively, especially when the accelerated factor is high. Besides, JDSI also owns nice robustness to abnormal subjects. |
| 3711 | Computer 131
|
PIFN EPT: MR-Based Electrical Property Tomography Using Physics-Informed Fourier Networks |
| Xinling Yu1, Jose Serralles2, Ilias Giannakopoulos3,4, Ziyue Liu5, Luca Daniel2, Riccardo Lattanzi3,4, and Zheng Zhang1 | ||
1Department of Electrical and Computer Engineering, University of California, Santa Barbara, Santa Barbara, CA, United States, 2Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, United States, 3Center for Advanced Imaging Innovation and Research (CAI2R), Department of Radiology, New York University Grossman School of Medicine, New York, NY, United States, 4Bernard and Irene Schwartz Center for Biomedical Imaging (CBI), Department of Radiology, New York University Grossman School of Medicine, New York, NY, United States, 5Department of Statistics and Applied Probability, University of California, Santa Barbara, Santa Barbara, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Electromagnetic Tissue Properties We introduce physics-informed Fourier networks (PIFNs) for Electrical Properties (EP) Tomography (EPT). Our novel deep learning-based method is capable of learning EPs globally from noisy magnetic resonance (MR) measurements, i.e, the magnitude of the magnetic transmit field and the transceive phase. Our proposed method also provides noise-free transmit field reconstructions. Two separate Fourier neural networks are used to efficiently estimate the transmit field and EPs at any location. We show that PIFN EPT accurately infers the EPs distribution of an inhomogeneous phantom from noisy simulated measurements. |
| 3712 | Computer 132
|
Super resolution imaging from low-field strength scanners using generative adversarial networks |
| Thomas Campbell Arnold1, Serhat V Okar2, Danni Tu3, Govind Nair2, John T. Pitts4, Megan E. Poorman4, Karan D. Kawatra2, Lisa M. Desiderio5, Matthew K. Schindler6, Brian Litt6, Russell T. Shinohara3, Daniel S. Reich2, and Joel M. Stein5 | ||
1Bioengineering, University of Pennsylvania, Philadelphia, PA, United States, 2National Institute of Neurological Disorders and Stroke, Bethesda, MD, United States, 3Biostatistics, University of Pennsylvania, Philadelphia, PA, United States, 4Hyperfine, Guilford, CT, United States, 5Radiology, University of Pennsylvania, Philadelphia, PA, United States, 6Neurology, University of Pennsylvania, Philadelphia, PA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Low-Field MRI, super resolution High-field MRI provides superior imaging for diverse clinical applications, but cost and other factors limit availability in various healthcare and lower resource settings. Lower-field strength units promise to expand access but involve tradeoffs including reduced signal, longer scan times, and lower resolution. Here we develop super-resolution methods that can generate high-field quality images from low-field scanner inputs, thus increasing signal and resolution. We use generative adversarial networks to demonstrate image enhancement in T1, T2 and FLAIR sequences. |
| 3713 | Computer 133
|
Impact of sampling strategies and residual U-net reconstruction on preserving high spatial frequencies in accelerated low-field MRI |
| Reina Ayde1,2, Tobias Senft1, Marco Fiorito1, Mauro Spreiter1, Najat Salameh1,2, and Mathieu Sarracanie1,2 | ||
1Center for Adaptable MRI Technology (AMT center), Department of Biomedical Engineering, University of Basel, Allschwil, Switzerland, 2AMT center, Institute of Medical Sciences, School of Medicine, Medical Sciences & Nutrition, University of Aberdeen, Aberdeen, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction, undersampling, averaging, data sampling Low signal-to-noise (SNR) ratios inherent to low-field (LF) MRI challenge its relevance in clinical applications. Accelerating the acquisition by undersampling k-space followed by reconstruction techniques has already shown promising results. Yet, undersampling is usually done by skipping high-frequency information which can lead to misdiagnosis as small lesions can be missed. In this study, we exploited a specificity of low-SNR regimes, that is signal averaging, to explore different acceleration strategies without skipping crucial information in k-space. The DL-reconstructed images arising from those sampling schemes have been evaluated on acquired in-vivo and ex-vivo LF-MRI datasets, showcasing high-frequency preservation and potential for generalization. |
| 3714 | Computer 134
|
A deep learning approach for compressed sensing reconstruction using adaptive shrinkage threshold |
| Yuan Lian1 and Hua Guo1 | ||
1Center for Biomedical Imaging Research, Department of Biomedical Engineering, School of Medicine, Tsinghua University, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Compressed Sensing is widely used for accelerating acquisitions of MR images. Deep learning methods have been introduced to CS-MRI reconstruction to improve image quality and computation speed. Here we introduce a new shrinkage function with adaptive threshold selection for Model-driven deep learning networks to suppress aliasing artifacts by utilizing the information on each feature map. We combine our adaptive threshold selection module with ISTA-Net, and demonstrate that the method can reduce reconstruction errors while preserving structural details effectively. |
| 3715 | Computer 135
|
The role of training on the robustness of domain-transform manifold learning |
| Danyal Bhutto1,2, Bo Zhu2, Jeremiah Zhe Liu3,4, Stephen Cauley2,5, Neha Koonjoo2,5, Bruce R Rosen2,5, and Matthew S Rosen2,5,6 | ||
1Biomedical Engineering, Boston University, Boston, MA, United States, 2Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, United States, 3Google, Mountain View, CA, United States, 4Biostatistics, Harvard University, Cambridge, MA, United States, 5Harvard Medical School, Boston, MA, United States, 6Physics, Harvard University, Boston, MA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Domain-transform manifold learning is a trained reconstruction approach where care needs to be taken to appropriately represent the forward encoding model during training, including for example the numerical properties of the source sensor data, phase relationship of complex sensor data, and field-of-view to prevent artifacts arising in the reconstruction. Here, we study the role that the training corpus and the numeral properties of the training have on the performance of the reconstruction of MRI data and demonstrate reconstruction artifacts that result from inference on out-of-training-distribution data if the training data is not augmented sufficiently. |
| 3716 | Computer 136
|
Relative noise variation with Unrolled Neural Networks for Accelerated Cardiac Cine Reconstruction |
| Suryanarayanan Sivaram Kaushik1, Xucheng Zhu2, Robert Marc Lebel3, Kavitha Manickam1, Ke Li1, Kailash Saravanan1, Florian Wiesinger4, and Martin Janich4 | ||
1GE Healthcare, Waukesha, WI, United States, 2GE Healthcare, Menlo Park, CA, United States, 3GE Healthcare, Calgary, AB, Canada, 4GE Healthcare, Munich, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Cardiovascular Deep Learning (DL) based reconstructions help alleviate the longer scan times seen in bSSFP Cine acquisitions by offering higher acceleration factors. This work analyzes the spatial and temporal variations in the noise as a function of acceleration factors in DL reconstructions of highly accelerated bSSFP Cine data. |
| 3717 | Computer 137
|
A noise robust image reconstruction deep neural network with cycle interpolation |
| Jeewon Kim1, Wonil Lee1, Beomgu Kang1, Seohee So2, and HyunWook Park1 | ||
1Korea Advanced Institute of Science and Technology (KAIST), Daejeon, Korea, Republic of, 2Korea Institute of Science and Technology, Seoul, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Parallel Imaging We propose a new Parallel Imaging scheme using a deep neural network which performs well with fewer ACS line in noisy environments. The proposed scheme includes ACS loss which is used in RAKI and cycle interpolation loss that we newly propose in our work. RAKI generalized GRAPPA in noisy environments by applying non-linear k-space interpolation with a deep neural network. However, it requires additional ACS lines to output satisfactory performance. Here, we suggest a new scheme to overcome the reconstruction performance in a noisy environment with fewer ACS lines. |
| 3718 | Computer 138
|
Feasibility of Deep Learning Reconstruction in the Clinical Application of MRI for patients with Bladder Cancer： a preliminary prospective study |
| Xinxin Zhang1, Yichen Wang1, Sicong Wang2, Min Li2, Yan Chen1, and Xinming Zhao1 | ||
1National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, 100021, China, Beijing, China, 2GE Healthcare, MR Research China, Beijing, Beijing, China |
||
Keywords: Urogenital, Bladder, Deep Learning Reconstruction The application of DLR significantly shortened scan times and improved the overall image quality score and image artifacts score and SNR and CNR of FSE-T2WI. DLR fast FSE-T2WI demonstrated significantly higher SNR (256.7±102.9 VS 94.7±40.8, p < 0.05) and CNR (168.0±77.3 VS 59.6±29.8, p < 0.05) and overall image quality scores (median, 4.0 vs. 3.0 for reader1 and 4.0 vs. 3.5 for reader2) than those of conventional FSE-T2WI. DLR may be useful in reducing the acquisition time of bladder MRI without compromising image quality. |
| 3719 | Computer 139
|
Rapid multi-slice whole-brain B1+-mapping at 7T using deep learning |
| Felix Krueger1, Christoph Stefan Aigner1, Max Lutz1, Layla Tabea Riemann1, Katja Degenhardt1, Bernd Ittermann1, Tobias Schaeffter1,2, Kerstin Hammernik3,4, and Sebasian Schmitter1,5,6 | ||
1Physikalisch-Technische Bundesanstalt, Berlin and Braunschweig, Germany, 2Division of Imaging Sciences and Biomedical Engineering, King's College London, London, United Kingdom, 3Technical University of Munich, Munich, Germany, 4Imperial College London, London, United Kingdom, 5Medical Physics in Radiology, German Cancer Research Center (DKFZ), Heidelberg, Germany, 6Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States |
||
Keywords: RF Pulse Design & Fields, Parallel Transmit & Multiband In this work we utilize deep learning to estimate multi-slice whole-brain B1+-maps in sub-seconds from initial localizer scans at 7T. The investigated neural networks use the receive profiles of the individual coil elements of an 8Tx/8Rx transceiver head coil as input information. The networks are trained on seven volunteers and tested in 2 unseen subjects for transversal/coronal/sagittal slices by comparing the prediction with the acquired B1+-maps. Subsequently, the feasibility of using the DL-based B1+-maps in a subject-specific calibration pipeline is demonstrated. |
| 3720 | Computer 140
|
Universal sequence-invariant deep learning image reconstruction for cardiovascular MR Multitasking |
| Zheyuan Hu1,2, Zihao Chen1,2, Hsu-Lei Lee1, Yibin Xie1, Debiao Li1,2, and Anthony Christodoulou1,2 | ||
1Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, United States, 2Department of Bioengineering, University of California, Los Angeles, Los Angeles, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Imaging Cardiovascular MR (CMR) Multitasking can quantify various parameter combinations without breath-holding or ECG monitoring. Clinically practical reconstruction time is viable using supervised deep subspace learning, but it depends on sequence-specific training. Here we explore whether universal, sequence-invariant CMR Multitasking deep learning reconstruction is practical by trading temporal awareness (breadth) for added depth in spatial domain. We evaluated the performance and generalizability of both strategies by training on T1 mapping data only and testing on two datasets: a) a matched-sequence T1 mapping data; and b) a novel-sequence T1-T2 mapping data. |
| 3875 | Computer 121
|
Estimating Uncertainty of Deep Learning for Tomographic Image Reconstruction Through Local Lipschitz |
| Danyal Bhutto1,2, Bo Zhu2, Jeremiah Zhe Liu3,4, Neha Koonjoo2,5, Bruce R Rosen2,5, and Matthew S Rosen2,5,6 | ||
1Biomedical Engineering, Boston University, Boston, MA, United States, 2Athinoula A. Martinos Center for Biomedical Imaging, Charlestown, MA, United States, 3Google, Mountain View, CA, United States, 4Biostatistics, Harvard University, Cambridge, MA, United States, 5Harvard Medical School, Boston, MA, United States, 6Physics, Harvard University, Boston, MA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction As deep learning approaches for image reconstruction become increasingly used in the radiological space, strategies to estimate reconstruction uncertainties become critically important to ensure images remain diagnostic. We estimate reconstruction uncertainty through calculation of the Local Lipschitz value, demonstrate a monotonic relationship between the Local Lipschitz and Mean Absolute Error, and show how a threshold can determine whether the deep learning technique was accurate or if an alternative technique should be employed. We also show how our technique can be used to identify out-of-distribution test images and outperforms baseline metrics, i.e. deep ensemble and Monte-Carlo dropout. |
| 3876 | Computer 122
|
Fast Uncertainty Estimation of IVIM parameters using Bayesian Neural Networks |
| Zehuan Zhang1, Matej Genči1, Hongxiang Fan1, Wayne Luk1, and Andreas Wetscherek2 | ||
1Department of Computing, Imperial College London, London, United Kingdom, 2Joint Department of Physics, The Institute of Cancer Research, London, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Diffusion/other diffusion imaging techniques, Uncertainty estimation We transformed the state-of-the art IVIM-NET for IVIM parameter fitting into a Bayesian Neural Network (BNN). BNNs can estimate uncertainty for quantitative MRI parameters, which is relevant for clinical decision making. We found that training on data with the highest SNR outperformed IVIM-BNNET models trained on matching SNR regarding parameter errors and uncertainties. A region with artificially increased noise could be identified from IVIM-BNNET's uncertainty output. Compared with traditional fitting, IVIM-BNNET achieved comparable accuracy, while being 21 times faster and providing less correlated parameter estimates. Monte-Carlo dropout rate 0.4 provided the best trade-off between low errors and low uncertainty. |
| 3877 | Computer 123
|
Super-Resolution Reconstruction of CEST images for human brain at 3T based on deep learning |
| Sirui Wu1, WenXuan Chen1, Yibing Chen2, Zhongsen Li1, and Xiaolei Song1 | ||
1Center for Biomedical Imaging Research, Department of Biomedical Engineering, Tsinghua University, Beijing 100084, China, Beijing, China, 2Xi’an Key Lab of Radiomics and Intelligent Perception, School of Information Sciences and Technology, Northwest University, Xi’an, Shaanxi 710069, China, Xi'an, China |
||
Keywords: Machine Learning/Artificial Intelligence, CEST & MT, Super-resolution, Unrolled network, quantitation CEST MRI suffers from the low resolution, leading to the disability to image small structures. We proposed a new optimization process adapting the CEST data form. Based on deep learning, our optimization flow is unfolded into an unrolled network for super resolution, termed as SC-Net. Besides, we introduced ROI-based normalization loss to guarantee quantitation accuracy. Results showed that SC-Net could reconstruct high-resolution image series and quantitative maps (PSNR = 44.27dB, CNR APTw = 41.79dB), when down sampling rate = 8. And the ROI-based normalization loss could calibrate errors. In conclusion, SC-Net has the potential to image small regions and acceleration. |
| 3878 | Computer 124
|
Investigation of deep learning reconstruction in MR imaging of pancreatic space occupying lesions |
| Meng Zhang1, Zheng Ye1, Bo Zhang2, Miaoqi Zhang2, and Zhenlin Li1 | ||
1Department of Radiology, West China Hospital, Sichuan University, Chengdu, China, 2MR Research, GE Healthcare, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction At present, the resolution of magnetic resonance image is still limited to the detection of pancreatic space occupying lesions. In this work, quantitative and qualitative analysis of pancreatic images of T1WI and DWI with built-in DL Recon was carried out. The results showed that the images obtained by the sequence with built-in DL Recon were better than the original images in terms of SNR, CNR and subjective scores. It is confirmed that deep learning reconstruction can improve image resolution and has potential in the detection of space occupying lesions in pancreas. |
| 3879 | Computer 125
|
Accurate Estimation of Background Phase in Virtual Conjugate Coil Expansion Combined with Wave Encoding |
| Congcong Liu1, Zhuoxu Cui1, Sen Jia1, Zhilang Qiu2, Xin Liu1, Hairong Zheng1, Dong Liang1, and Haifeng Wang1 | ||
1Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, 2Case Western Reserve University, Cleveland, OH, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction The wave encoding model with virtual conjugate coil (Wave-VCC) extension can provide more powerful MRI-accelerated imaging performance. However, estimating the background phase covering the full frequency range is un-tractable by only acquiring the middle auto-calibration signals (ACS) line during reconstruction. Here, combining a neural network without training, a new method to generate more accurate background phase in Wave-VCC is proposed. Including ablation experiments and comparison, experiments were carried out to verify the feasibility and performance of the proposed methods, respectively. |
| 3880 | Computer 126
|
Deep Learning Reconstruction of Dynamic Free-breathing Fetal Heart MRI to Improve Clinical Pipeline |
| Denis Prokopenko1, Daniel Rueckert2,3, and Joseph V. Hajnal1 | ||
1Biomedical Engineering Department, School of Biomedical Engineering and Imaging Sciences, King's College London, London, United Kingdom, 2Department of Informatics, Technical University of Munich, Munich, Germany, 3Department of Computing, Imperial College London, London, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Dynamic free-breathing fetal heart MRI requires high spatial and temporal resolution, which could be reconstructed by kt-SENSE from undersampled data guided by priors of the same anatomy. Doubled acquisition time and uncontrolled fetal motion between the 2 acquisitions affects the data quality for reconstruction. We explored an alternative deep learning approach using a 3D U-Net based model with time-averaged skip connection and data consistency. Assessment of the model set a baseline for prior preconstruction and underlines important pitfalls that will drive further improvements to achieve optimal reconstruction quality. |
| 3881 | Computer 127
|
Uncertainty Estimation for Deep Learning-Based Enhancement of Undersampled Dual-Echo Steady-State Knee MRI |
| Ashmita Deb1,2, Shu-Fu Shih1,2, Zhaohuan Zhang1,2, and Holden H Wu1,2 | ||
1Department of Radiological Sciences, David Geffen School of Medicine, University of California Los Angeles, Los Angeles, CA, United States, 2Department of Bioengineering, University of California Los Angeles, Los Angeles, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Data Analysis, Image Enhancement Deep learning (DL) based image enhancement of undersampled 3D dual-echo steady-state knee MRI can achieve faster computation times compared to compressed sensing reconstruction. However, it is hard to interpret how DL models work. This introduces the risk of DL-enhanced images containing inaccuracies without the user’s knowledge and thus confounding diagnosis. This work aimed to calculate pixel-wise uncertainty maps for DL-enhanced images by incorporating Monte Carlo Dropout into a 2D UNET to estimate epistemic uncertainty. Analysis showed that the DL-enhanced images achieved good image quality and the spatial uncertainty maps reflected errors, compared to reference images. |
| 3882 | Computer 128
|
Accelerating Phase and Quantitative susceptibility mapping with Scan-Specific Complex Convolutional Neural Networks |
| Swetali Nimje1,2, Ludovic de Rochefort1, and Thierry Artières2 | ||
1Aix Marseille University, CNRS, CRMBM, Marseille, France, 2Aix Marseille University, Ecole Centrale de Marseille, CNRS, LIS, Marseille, France |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Susceptibility mapping MRI data is inherently complex-valued, the vast majority of deep learning frameworks do not yet support complex-valued data. Most reconstruction networks separate real and imaginary components into two separate real-valued channels, which may not be the most efficient way to represent complex numbers. Phase is essential for many MRI applications, including phase contrast velocity mapping and Quantitative Susceptibility Mapping (QSM) etc. We propose a new crRAKI, a scan-specific complex-valued residual convolutional neural network for 2D/3D MRI data for accelerating phase mapping and QSM. A comparison is made with GRAPPA and rRAKI for the accelerated reconstruction of MRI images. |
| 3883 | Computer 129
|
Evaluation of deep learning-based reconstruction for qualitative and quantitative DW-MRI in head and neck cancers |
| Ramesh Paudyal1, Akash Deelip Shah2, Amaresha Shridhar Konar1, Jaemin Shin3, Eve LoCastro1, Nisha Bagchi4, Maggie Fung3, Suchandrima Banerjee5, Nancy Lee6, and Amita Shukla-Dave1,2 | ||
1Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, United States, 2Radiology, Memorial Sloan Kettering Cancer Center, New York, NY, United States, 3GE Health Care, New York, NY, United States, 4Kimmel Medical College, Thomas Jefferson University, Philadelphia, PA, United States, 5GE Health Care, Menlo Park, CA, United States, 6Radiation Oncology, Memorial Sloan Kettering Cancer Center, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Tumor The head and neck (HN) region have complex anatomical structures that affect the image quality of diffusion-weighted MRI. Therefore deep learning (DL)-based Reconstruction (Recon) for DW-MRI could be a promising method that can help improve image sharpness and signal-to-noise ratio (SNR) without increasing signal averaging. The present study aimed to evaluate the performance of qualitative and quantitative multiple b-value DW-MRI powered by DL-based Recon for tumors in the HN region. The DL-based recon method improved the DW image quality and SNR compared to those without DL recon. |
| 3884 | Computer 130
|
Implicit CINE: a deep-learning super-resolution model for multi-planar real-time MRI |
| Nora Vogt1, Karyna Isaieva1, Jean-Sébastien Louis1, Christian Weihsbach2, Mattias Paul Heinrich2, Freddy Odille1,3, Pierre-André Vuissoz1,3, Jacques Felblinger1,3, and Julien Oster1,3 | ||
1IADI, Université de Lorraine, INSERM U1254, Nancy, France, 2Institut für Medizinische Informatik, Universität zu Lübeck, Lübeck, Germany, 3CIC-IT, INSERM 1433, Université de Lorraine and CHRU Nancy, Nancy, France |
||
Keywords: Machine Learning/Artificial Intelligence, Cardiovascular Cardiac CINE MRI plays an essential role in cardiac diagnosis, but not all patients are eligible for 3D imaging, which is associated with long acquisition times. We propose a deep-learning super-resolution model to generate 3D CINE from multi-planar 2D real-time MRI using external signals for cardiac and respiratory motion estimation. The proposed neural field model is trained on a single subject and performs non-rigid motion compensation and implicit representation learning in an end-to-end manner. A preliminary study with three healthy volunteers demonstrates promising reconstruction performance and computation times compared to traditional registration-based approaches. |
| 3885 | Computer 131
|
Accelerated MRI using Dual-domain Transformer-Based Reconstruction and Learning-based Undersampling |
| Guan Qiu Hong1,2,3, William Morley2, Matthew Wan2, Yuan Tao Wei1,2,3, Yang Su2, and Hai-Ling Margaret Cheng1,2,3 | ||
1Institute of Biomedical Engineering, University of Toronto, Toronto, ON, Canada, 2The Edward S. Rogers Sr. Department of Electrical and Computer Engineering, University of Toronto, Toronto, ON, Canada, 3Translational Biology & Engineering Program, Ted Rogers Centre for Heart Research, Toronto, ON, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Deep learning architectures such as Convolutional Neural Networks (CNNs) are widely used alongside fixed k-space undersampling for reconstructing highly undersampled MRI k-space data. However, CNN-based architectures may perform sub-optimally due to their limited ability to capture long range dependencies, and fixed undersampling patterns may not be amenable to optimal reconstruction. We propose dual-domain (image and k-space), transformer-based reconstruction architectures paired with learning-based undersampling to reconstruct undersampled k-space data. Experimental results demonstrate improved reconstruction quality compared to CNN-based (UNet) architecture across 5x to 100x acceleration factors; dual domain outperformed single domain reconstructions. |
| 3886 | Computer 132
|
Data-adapted Neural Network Denoisers as a Regularization Engine for Low-latency Image Reconstruction in Accelerated Cardiac Perfusion MRI |
| Dilek Mirgun Yalcinkaya1,2, Hazar Benan Unal1, Subha Raman3,4, Abolfazl Hashemi2, Rohan Dharmakumar3,4, and Behzad Sharif1,3,4 | ||
1Laboratory for Translational Imaging of Microcirculation, Indiana University (IU) School of Medicine, Indianapolis, IN, United States, 2Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, United States, 3Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, United States, 4Krannert Cardiovascular Research Center, IU School of Medicine/IU Health Cardiovascular Institute, Indianapolis, IN, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction, Plug-and-play, denoiser In this work, we demonstrated that a deep-learning based denoiser trained on a limited dataset of first-pass myocardial perfusion MRI studies enables low-latency image reconstruction using the plug-and-play iterative reconstruction framework. Our proof-of-concept results suggest that the data-adapted denoiser resulted in superior performance versus a generic denoiser especially if there are constraints on the number of iterations (total computation time) which is the case in certain clinical settings specifically interventional MRI. Our findings also imply that radial sampling may be a more desirable data acquisition strategy for PnP-based image reconstruction in first-pass MP MRI studies. |
| 3887 | Computer 133
|
MRI Denoising with a Non-Blind Deep Complex-Valued Convolutional Neural Network |
| Quan Dou1, Zhixing Wang1, Xue Feng1, and Craig Meyer1 | ||
1Biomedical Engineering, University of Virginia, Charlottesville, VA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Signal-to-noise ratio is crucial for MR image analysis. In this work, we proposed a non-blind complex-valued convolutional neural network for MRI denoising. Compared to existing denoising methods, the proposed method shows better performance on handling noise at different levels and on spatially variant noise. |
| 3888 | Computer 134
|
Deep Learning Reconstruction for Magnetic Resonance Image Quality Improvement in Lumbar Endplate Inflammation |
| Xinyang Lv1, Zheng Ye1, Miaoqi Zhang2, Bo Zhang2, and Zhenlin Li1 | ||
1radiology department, West China Hospital，Sichuan University, Chengdu, China, 2MR Research, GE Healthcare, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Magnetic resonance imaging (MRI) is a useful tool to diagnose lumbar endplate inflammation. It is thus important to improve diagnostic accuracy by improving image quality. In this study, we compared signal-to-noise ratio (SNR), contrast noise ratio (CNR) and subjective scores between original images and deep learning reconstruction (DL Recon) images in 31 patients diagnosed with lumbar endplate inflammation. It was observed that the deep learning reconstructed images outperformed conventional images in terms of both subjective scores and objective values. |
| 3889 | Computer 135
|
Reconstruction of accelerated MR cholangiopancreatography using supervised and self-supervised 3D Variational Networks |
| Jonas Kleineisel1, Bernhard Petritsch1, Thorsten A. Bley1, Herbert Köstler1, and Tobias Wech1 | ||
1Department of Diagnostic and Interventional Radiology, University Hospital of Würzburg, Würzburg, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Magnetic resonance cholangiopancreatography suffers from long examination times and artifacts originating from residual motion. To shorten the acquisition protocol, we acquired data at 12-fold undersampling and investigated a 3D Variational Network (VN) architecture for reconstruction. We compared a self-supervised training scheme and a supervised network trained on synthetic data. We find that the self-supervised method is only able to provide competitive reconstructions if the network is initialized with pre-trained weights, and even then does not offer superior performance over the supervised approach. For the presented exemplary data, the supervised VN showed comparable image quality as a reference Compressed Sensing model. |
| 3890 | Computer 136
|
Deep Learning Reconstruction in MRI: Comparison of Image Quality in Patients With Hepatic Malignancy |
| Yuqi Tan1, Zheng Ye1, Miaoqi Zhang2, Bo Zhang2, and Zhenlin Li1 | ||
1West China Hospital of Sichuan University, Chengdu, China, 2GE Healthcare, MR Research, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Improving image resolution by denoising is an important research goal in MRI reconstruction. An emerging technique, deep learning reconstruction (DLR), has shown great potential in MRI denoising. In this study, we included 37 patients with pathologically diagnosed hepatic malignancy, and compared the image quality of DLR and original reconstruction regarding dual echo T1 weighted sequence, diffusion weighted imaging (DWI) and fat-suppressed T1 weighted gadolinium-enhancement. It was shown that DLR significantly improved the image quality by reducing background noise, thus making hepatic malignancy more conspicuous. |
| 3891 | Computer 137
|
Overview of Complex-valued Image Reconstruction for CS-MRI Using Real-valued CNN with Symmetrical Signal Under-Sampling |
| Shohei Ouchi1, Itona Fukatsu1, Kazuki Yamato1, and Satoshi Ito1 | ||
1Utsunomiya University, Utsunomiya, Japan |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Complex-valued CNN based image reconstruction methods have been proposed to correspond to MR images with a spatial phase variation. However, using those CNN may lead to over-fitting because CNN layers for complex numbers are requires large number of parameters than real-valued CNN. We previously proposed a reconstruction method for complex-valued image using a real-valued DnCNN by introducing a symmetrical k-space under-sampling. In this study, we introduced this method to U-Net and ADMM-CSNet. Reconstruction experiments showed that a real-valued CNN has the possibility to have the same or better performance as a complex-valued CNN without perform complex calculations. |
| 3892 | Computer 138
|
A Two-Stage Super-Resolution (TSSR) CEST Model Using Deep-Learning Reconstruction |
| Wenxuan Chen1, Sirui Wu1, and Xiaolei Song1 | ||
1Center for Biomedical Imaging Research, Department of Biomedical Engineering, Tsinghua University, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Super-resolution CEST is an important source of new contrast in MRI. However, it is time-consuming to obtain high-resolution CEST images. We proposed a deep learning based Two-Stage Super-Resolution (TSSR) model for CEST images. Compared with conventional SISR or MFSR models, TSSR model can effectively utilize the correlations among slices and those among saturation offsets for CEST images. We acquired brain CEST images on 14 volunteers using a 3T clinical MR scanner. Initial results suggested that the proposed TSSR model outperformed other methods for all the evaluation indicators. Our work showed the potential in reconstruction of high-resolution CEST images from low-resolution ones. |
| 3893 | Computer 139
|
Joint Estimation of Coil Sensitivity and Image by Using Untrained Neural Network without External Training Data |
| Gulfam Ahmed Saju1, Zhiqiang Li2, Reza Abiri3, Tianming Liu4, and Yuchou Chang1 | ||
1Computer and Information Science, University of Massachusetts Dartmouth, North Dartmouth, MA, United States, 2Neuroradiology, Barrow Neurological Institute, Phoenix, AZ, United States, 3Electrical, Computer and Biomedical Engineering, University of Rhode Island, Kingston, RI, United States, 4Computer Science, University of Georgia, Athens, GA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Parallel Imaging Training data for MRI reconstruction are difficult to be acquired in clinical practice. In addition, machine learning or deep learning-based MRI reconstruction suffers the distribution shift problem between training data and testing data. Generalization error always exists, so reconstructed images are unstable. We proposed a joint estimation of coil sensitivity and image using the prior of an untrained neural network (UNN). Coil sensitivity map improvement gradually enhances the UNN prior and the image to be reconstructed in an iterative optimization process. The method outperforms other MRI reconstruction methods by suppressing noise and aliasing artifacts. |
| 3894 | Computer 140
|
Deep learning-based Motion-corrected Rapid Image Reconstruction for High-resolution Cartesian First-pass Myocardial Perfusion Imaging at 3 T |
| Junyu Wang1 and Michael Salerno1 | ||
1Cardiovascular Medicine, Stanford University, Stanford, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Keywords: artificial intelligence, image reconstruction, perfusion
Cardiac magnetic resonance first-pass contrast-enhanced myocardial perfusion imaging is valuable for evaluating coronary artery disease1. 2D Cartesian perfusion imaging using compressed sensing (CS)-based reconstructions such as L1-SENSE2 enables fast and high-resolution imaging, but whole-heart coverage cannot be achieved without simultaneous multi-slice (SMS) acquisitions and the CS-based iterative reconstruction takes ~30 minutes per slice. To address this, we have developed a deep learning-based motion-corrected rapid image reconstruction for high-resolution Cartesian perfusion imaging at 3 Tesla, for both 2D and SMS MB=2 acquisitions, which provides fast and high-quality motion-corrected reconstruction and makes rapid online reconstruction feasible. |
| 4031 | Computer 101
|
Adaptive Deep Learning MR Image Enhancement with Property Constrained Unrolled Network |
| Zechen Zhou1, Ryan Chamberlain1, Praveen Gulaka1, Enhao Gong1, Greg Zaharchuk1, and Ajit Shankaranarayanan1 | ||
1Subtle Medical Inc, Menlo Park, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence An unrolled network with explicit MR image degradation modeling and physical property constraints, termed PGDNet, is proposed for adaptive image denoising and deblurring. Preliminary evaluation demonstrated that PGDNet can achieve similar/superior image quality enhancement compared to the conventional task-specific networks, and can outperform others in joint denoising and deblurring tasks. PGDNet provides a promising solution for adaptive MR image denoising and deblurring to restore the image quality of accelerated clinical MR scans. |
| 4032 | Computer 102
|
Performance Evaluation of Deep Learning-based Image Reconstruction for Head and Neck Imaging Protocol |
| Amaresha Shridhar Konar1, Jaemin Shin2, Ramesh Paudyal1, Abhay Dave3, Maggie Fung2, Suchandrima Banerjee4, Vaios Hatzoglou5, and Amita Shukla-Dave1,5 | ||
1Medical Physics, Memorial Sloan Kettering Cancer Center, New York City, NY, United States, 2GE Healthcare, New York City, NY, United States, 3Touro College of Osteopathic Medicine, New York, NY, United States, 4GE Healthcare, Menlo Park, CA, United States, 5Radiology, Memorial Sloan Kettering Cancer Center, New York City, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Imaging MRI has excellent extracranial soft-tissue contrast to detect tumors in the head and neck (HN) region. Technical challenges arise due to MRI related artifacts. In routine radiological practice, HN MR imaging protocols are optimized specifically to the subsites. We aimed to evaluate the performance of the HN imaging protocol that include qualitative T1w, T2w, and quantitative diffusion MRI powered by a novel deep learning (DL) based reconstruction (recon) using the ACR and QIBA diffusion phantoms. This phantom study showed that qualitative T1w and T2w images and multiple b-value DWI data powered with DL recon substantially improves the image quality. |
| 4033 | Computer 103
|
Image sequence partitioning in CRNN-based reconstruction improves ventilation defect detection of 3D-PREFUL with undersampled data |
| Maximilian Zubke1,2, Filip Klimeš1,2, Andreas Voskrebenzev1,2, Marius Wernz1,2, Till F Kaireit1,2, Agilo L Kern1,2, Marcel Gutberlet1,2, Robin A Müller1,2, Frank Wacker1,2, and Jens Vogel-Claussen1,2 | ||
1Institute of Diagnostic and Interventional Radiology, Hannover Medical School, Hannover, Germany, 2Biomedical Research in Endstage and Obstructive Lung Disease Hannover (BREATH), German Center for Lung Research (DZL), Hannover, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Lung 3D-phase-resolved-functional-lung (3D-PREFUL) MRI enables a non-contrast-enhanced detection of lung ventilation defects. Convolutional, recurrent neural networks (CRNN) can accelerate data acquisition by reconstructing dynamic lung images from undersampled data. However, a remaining bias in reconstruction led to incorrect detection of ventilation defects. Reducing the number of respiratory phases per reconstruction step improved the accuracy of the defect detection. This improvement is demonstrated by comparison of 2 complementary ventilation defect metrics derived from the original data and the 2x, 4x and 6x undersampled data of Asthma, COPD and post-COVID-19 patients reconstructed from a CRNN in partitions of 30,20 and 10 respiratory phases. |
| 4034 | Computer 104
|
DeepGraspT1: Deep Learning-Enabled GRASP T1 Mapping |
| Haoyang Pei1,2, Ding Xia1, Fang Liu3, and Li Feng1 | ||
1Biomedical Engineering and Imaging Institute and Department of Radiology, Icahn School of Medicine at Mount Sinai, New York, NY, United States, 2Department of Electrical and Computer Engineering, NYU Tandon School of Engineering, New York, NY, United States, 3Athinoula A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Boston, MA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Golden-angle RAdial Sparse Parallel (GRASP) MRI has recently been extended for rapid, accurate and robust T1 mapping (GraspT1) that can be performed during free breathing. However, GraspT1 implements a conventional T1 mapping framework that reconstructs an image series from undersampled dynamic k-space in the first step and then performs pixel-wise parameter fitting in the second step. This leads to a slow and cumbersome pipeline to obtain T1 maps. In this work, we developed deep learning-based GraspT1 (DeepGraspT1), which directly estimates T1 maps from undersampled k-space and enables additional acceleration that outperforms conventional iterative reconstruction. |
| 4035 | Computer 105
|
Exploring reproducibility in deep learning-based parallel imaging reconstruction |
| Chungseok Oh1, Hongjun An1, and Jongho Lee1 | ||
1Seoul National University, Seoul, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence The performance of a deep neural network can be affected by software and hardware setups when training the network and, therefore, can vary from training to training. This issue of reproducibility, which can be referred to as an “intrinsic” reproducibility of deep learning, can be critical for academic research because reproducibility is a key requirement for journal papers. In this study, we explore this intrinsic reproducibility issue for deep learning-powered parallel imaging reconstruction by using a popular end-to-end variational network. This study may provide minimal requirements for reproducible research in network training. |
| 4036 | Computer 106
|
Fouier Convolution Nerual Network for MRI reconstruction |
| Haozhong Sun1, Yuze Li1, Runyu Yang1, Zhongsen Li1, and Huijun Chen1 | ||
1Center for Biomedical Imaging Research, Tsinghua University, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Various image reconstruction methods have been proposed to reduce Magnetic resonance (MR) image acquisition time. One of recent trends is convolution neuron network (CNN) based deep learning model. However, most of these CNN models keep architecture of stacking small filters (e.g. 1×1 or 3×3) and the effective receptive field of these networks is limited, which is undesired for reconstruction because the random undersampling pattern causes global artifact. We proposed Fourier convolution block (FCB) to replace regular convolution filters. FCB can achieve both global receptive field and high computing efficiency by multiplication in frequency domain. |
| 4037 | Computer 107
|
Improving JSENSE Using an Initial Reconstruction with an Unrolled Deep Network Prior |
| Gulfam Ahmed Saju1, Zhiqiang Li2, Reza Abiri3, Tianming Liu4, and Yuchou Chang1 | ||
1Computer and Information Science, University of Massachusetts Dartmouth, North Dartmouth, MA, United States, 2Neuroradiology, Barrow Neurological Institute, Phoenix, AZ, United States, 3Electrical, Computer and Biomedical Engineering, University of Rhode Island, Kingston, RI, United States, 4Computer Science, University of Georgia, Athens, GA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Parallel Imaging JSENSE iteratively optimizes sensitivity maps and the image, so the sensitivity profile can be gradually improved during the reconstruction process. The initially reconstructed image in the first iteration is obtained by the initially estimated coil sensitivity maps. The initial coil sensitivity profiles may be inaccurate and therefore degrade the quality of the subsequent image quality and coil sensitivity map estimation in the iterative optimization process. We propose to use unrolled deep network prior to replace the initial reconstruction in the conventional JSENSE for improving the image reconstruction quality. Experimental results show that the proposed method outperforms CG-SENSE and JSENSE. |
| 4038 | Computer 108
|
Incorporating Untrained Neural Network Prior in PROPELLER Imaging |
| Gulfam Ahmed Saju1, Zhiqiang Li2, Reza Abiri3, Tianming Liu4, and Yuchou Chang1 | ||
1Computer and Information Science, University of Massachusetts Dartmouth, North Dartmouth, MA, United States, 2Neuroradiology, Barrow Neurological Institute, Phoenix, AZ, United States, 3Electrical, Computer and Biomedical Engineering, University of Rhode Island, Kingston, RI, United States, 4Computer Science, University of Georgia, Athens, GA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Motion Correction Periodically Rotated Overlapping ParallEL Lines with Enhanced Reconstruction (PROPELLER) MRI technique enables the correction of motion artifacts resulted from patient motions in a scanner. Undersampling the blades can increase data acquisition speed and reduce potential motions caused by pains in a short time but may degrade image quality. Deep neural networks may support the blade reconstruction with undersampled data but motion patterns are difficult to be acquired for building a training dataset. To avoid the acquisition of training data, this abstract proposes an untrained neural network-based PROPELLER reconstruction technique to enhance image quality with undersampled blades. |
| 4039 | Computer 109
|
MR reconstruction in k-space using Vision Transformer boosted with Masked Image Modeling |
| Jaa-Yeon Lee1 and Sung-Hong Park2 | ||
1Korea Advanced Institute of Science & Technology, Daejeon, Korea, Republic of, 2Korea Advanced Institute of Science & Technology, Daejoen, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Vision Transformer, Masked Image Modeling Masked image modeling (MIM) skim has recently been shown to pre-train vision transform (ViT) and works as an effective data augmentation. In this study, we proposed MR reconstruction algorithm using ViT with MIM in k-space. The proposed method showed better performance than the original 1D and 2D ViT. Our study showed that MIM can be used to enhance the reconstruction quality to help learn the data distribution and that combining the loss in both k-space and spatial domain reconstructs images with better perceptuality. |
| 4040 | Computer 110
|
Deep learning acceleration and reconstruction for single-channel signals |
| Yongquan Ye1, Zhongqi Zhang1, Eric Z Chen2, Xiao Chen2, Shanhui Sun2, and Jian Xu1 | ||
1United Imaging, Houston, TX, United States, 2United Imaging Intelligence, Cambridge, MA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction The capacity of the deep learning ReconNet3D model for single-channel image reconstruction with highly under-sampled k-spaces is demonstrated. Without the need for coil sensitivity information, the proposed method can achieve an acceleration factor of 4 on a dual-channel VTC coil. Supporting high-factor acceleration with limited coil channels can be very beneficial for imaging with surface coils or preclinical animal scans. |
| 4041 | Computer 111
|
Deep Learning Based Reconstruction Improved Image Quality for rectum T2-weighted imaging |
| Weiming Feng1, Lan Zhu1, Yihan Xia1, Kangning Wang1, Yong Zhang2, Jiankun Dai3, Guifeng Fu3, and Huan Zhang1 | ||
1Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai, China, 2GE Healthcare, Shanghai, China, 3GE Healthcare, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction, rectum; magnetic resonance imaging; T2-weighted imaging High-resolution MRI is of much significance in preoperatively staging rectal cancer. However, the motion artifact from intestinal peristalsis inevitably affects image quality then the accuracy of staging. Deep learning reconstruction (DLRecon) that uses artificial neural networks to extract patterns and makes predictions from large data sets, has been verified in related studies for improving image quality and reducing scanning time. In this study, rectum T2-weighted imaging (T2WI) reconstructed with DLRecon and conventional reconstruction were evaluated, and the results indicate that DLRecon could be employed for better image quality without extra scanning time in clinical practice. |
| 4042 | Computer 112
|
Quantitative Evaluation of Deep Learning Reconstruction of Diffusion-weighted MRI using a DWI Phantom |
| Xiangchuang Kong1, Shengzhen Tao1, Eric H. Middlebrooks1, Thomas Benkert2, Xiangzhi Zhou1, and Chen Lin1 | ||
1Department of Radiology, Mayo Clinic, Jacksonville, FL, United States, 2Siemens Medical Solutions USA, Inc., Jacksonville, FL, United States, JACKSONVILLE, FL, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Diffusion/other diffusion imaging techniques, deep learning reconstruction; DWI phantom In a quantitative phantom study, deep learning (DL) reconstruction is shown to improve the SNR ratio of DWI images while preserving measured ADC values. The average SNR gain is similar to that achieved with better gradient performance between Siemens Prisma and Vida. Such benefits of DL recon should allow better quality and/or shorter scanning of DWI in clinical applications. |
| 4043 | Computer 113
|
Deep Learning Reconstruction for 4-fold Accelerated 2DFSE Imaging: optimization of variable density undersampling |
| Michael Carl1, Rafi Brada2, Nir Mazor2, Daniel V Litwiller3, and Maggie Fung4 | ||
1GE Healthcare, San Diego, CA, United States, 2GE Research, Herzliya, Israel, 3GE Healthcare, Denver, CO, United States, 4GE Healthcare, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence In this work we use variable-density prospective undersampling of the phase-encode k-space lines (ky) in 2D fast spin-echo (2DFSE) followed by deep learning (DL) reconstruction. We were able to achieve an acceleration of R=4 while maintaining high image quality. |
| 4044 | Computer 114
|
Highly Accelerated Multi-channel 3D Knee Imaging Using Denoising Diffusion Probabilistic Model (DDPM) and GRAPPA |
| Ruiying Liu1, Jee Hun Kim2, Hongyu Li1, Peizhou Huang1, Xiaojuan Li2, and Leslie Ying1 | ||
1Department of Biomedical Engineering, Department of Electrical Engineering, University at Buffalo, Buffalo, NY, United States, 2Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, OH, United States |
||
Keywords: Image Reconstruction, Machine Learning/Artificial Intelligence Deep learning methods have achieved superior reconstruction results in MRI reconstruction. Recently, denoising diffusion probabilistic models (DDPM) have demonstrated great potential in image-processing tasks. In this study, we combine denoising diffusion probabilistic models (DDPM) and GRAPPA for highly accelerated 3D imaging. The method sequentially performs DDPM and GRAPPA with specially designed sampling masks such that the benefits of the diffusion model and the availability of multi-channel data can be utilized jointly. Our results demonstrate that the proposed method can achieve an acceleration factor of up to 16 which is the product of the factors achieved by DDPM and GRAPPA alone. |
| 4045 | Computer 115
|
Finding Optimal Regularization Parameter for Undersampled Reconstruction using Bayesian Optimization |
| Alberto Di Biase1,2, Claudia Prieto1,2, and Rene Botnar1,2 | ||
1Department of Electrical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile, 2Millennium Institute for Intelligent Healthcare Engineering (iHEALTH), Santiago, Chile |
||
Keywords: Image Reconstruction, Machine Learning/Artificial Intelligence, Compressed sensing We present a Bayesian optimization approach to find the optimal regularization parameters for undersampled MRI reconstructions such as Compressed Sensing (CS). Bayesian optimization can find optimal points for expensive to evaluate functions by efficient sampling the next points to try by maximizing the expected improvement. Additionally, the use of pruning can speed up optimization by stopping early unpromising trials. We demonstrate the effectiveness of this optimization technique by finding optimal parameters for undersampled MRI using Total Variation and CS-Wavelet regularization. The parameters found with the proposed approach are comparable to does found by grid search but requiring shorter computational times. |
| 4046 | Computer 116
|
Quantitative evaluation of denoising algorithms without noise-free ground-truth data |
| Laura Pfaff1,2, Fabian Wagner1, Julian Hossbach1,2, Elisabeth Preuhs1, Dominik Nickel2, Tobias Wuerfl2, and Andreas Maier1 | ||
1Pattern Recognition Lab, Friedrich-Alexander-Universität Erlangen-Nürnberg, Erlangen, Germany, 2Magnetic Resonance, Siemens Healthcare GmbH, Erlangen, Germany |
||
Keywords: Image Reconstruction, Machine Learning/Artificial Intelligence In MRI, the quantitative evaluation of denoising methods is often limited due to the lack of noise-free ground-truth data. We show how to still approximate the quality metrics mean squared error (MSE) and peak signal-to-noise-ratio (PSNR) without access to ground-truth data by using Stein’s unbiased risk estimator (SURE). The proposed method can be employed to evaluate learning- and non-learning-based denoising approaches, assuming an additive Gaussian noise model with known distribution. Our experiments further reveal that the accuracy of our evaluation method increases with the number of test samples available. |
| 4047 | Computer 117
|
Impact of Deep Learning-Based MR Image Reconstruction Algorithms on Abdominal MRI Radiomic Features |
| Hailong Li1, Vinicius Vieira Alves1, Amol Pednekar1, Mary Kate Manhard1, Joshua Greer1, Andrew T. Trout1, Lili He1, and Jonathan R. Dillman1 | ||
1Cincinnati Children's Hospital Medical Center, Cincinnati, OH, United States |
||
Keywords: Image Reconstruction, Machine Learning/Artificial Intelligence, deep learning reconstruction Deep learning (DL)-based techniques are increasingly being applied to assist with or improve image reconstruction. However, the impact of DL-based algorithms on radiomics is not well understood. This study aims to evaluate the impact of two commercially available DL-based reconstruction pipelines: (1) SmartSpeed (Philips Healthcare, U.S. FDA-cleared); and (2) SmartSpeed with Super Resolution (SmartSpeed+SuperRes, not U.S. FDA-cleared to date) on MRI radiomic features. Our analysis showed that compared to conventional image reconstruction technique, 42 out of 86 investigated radiomic features from SmartSpeed images were highly correlated whereas only 13 features from SmartSpeed+SuperRes images had high correlations with conventional image features. |
| 4048 | Computer 118
|
AI assisted upscaling of low-resolution cardiac MRI reduces acquisition time and yields qualitatively comparable images to standard techniques |
| Dmitrij Kravchenko1,2, Alexander Isaak1,2, Narine Mesropyan1,2, Claus Christian Pieper1, Daniel Kuetting1,2, Leon M. Bischoff1,2, Shuo Zhang3, Christoph Katemann3, Johannes M. Peeters4, Oliver Weber3, Ulrike Attenberger1, and Julian Luetkens1,2 | ||
1Diagnostic and interventional radiology, University Hospital Bonn, Bonn, Germany, 2Quantitative Imaging Laboratory Bonn, Bonn, Germany, 3Philips GmbH Market DACH, Hamburg, Germany, 4Philips MR Clinical Science, Best, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Cardiovascular AI assisted upscaling of low-resolution cine bSSFP images yields comparable image quality to conventional images with no clinically significant difference in volumetric data at a reduction of acquisition time by a factor of 1.5 to 2. Keywords: Artificial intelligence, acceleration, Superresolution, cardiac MRI |
| 4049 | Computer 119
|
Self-Supervised Denoising for Longitudinal MRI |
| Matt Hemsley1,2, Liam S.P Lawrence1,2, and Angus Z Lau1,2 | ||
1Medical Biophysics, University of Toronto, Toronto, ON, Canada, 2Physical Sciences Platform, Sunnybrook Research Institute, Toronto, ON, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Cancer, Denoising, Self-Supervised In MR guided radiation therapy, images of patients are acquired daily. However, scan times are long to acquire images with an acceptable SNR for treatment planning and adaptation. In this study we test a self-supervised machine learning based approach for denoising data that can utilize previous scans of the same patient to improve quality of the denoised image. Results on a numerical phantom and clinical images are presented and compared to a popular non-machine learning denoising algorithm. |
| 4050 | Computer 120
|
Safeguarded Deep Unfolding Network for Parallel MR Imaging |
| Zhuo-Xu Cui1, Sen Jia2, Jing Cheng2, Qingyong Zhu1, and Dong Liang1,3 | ||
1Medical AI Research Center, Shenzhen Institutes of Advanced Technology, Chinese Academy of Science, Shenzhen, China, 2Paul C. Lauterbur Research Center for Biomedical Imaging, Shenzhen Institutes of Advanced Technology, Chinese Academy of Science, Shenzhen, China, 3Pazhou Lab, Guangzhou, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction This study proposes a safeguarded methodology for network unrolling. Specifically, focusing on parallel MR imaging, we unroll a zeroth-order algorithm, of which the network module represents a regularizer itself so that the network output can still be covered by a regularization model. Furthermore, inspired by the idea of deep equilibrium models, before backpropagation, we carry out the unrolled network to converge to a fixed point and then prove that it can tightly approximate the real MR image. In a case where the measurement data contain noise, we prove that the proposed network is robust against noisy interferences. |
| 4187 | Computer 101
|
Improving Brain Volume Measurement Workflow using combination of CS-MRI and Deep Learning based Super-Resolution |
| Atita Suwannasak1, Uten Yarach1, and Prapatsorn Sangpin2 | ||
1Chiang Mai university, Chiang Mai, Thailand, 2Philips Healthcare (Thailand), Bangkok, Thailand |
||
Keywords: Machine Learning/Artificial Intelligence, Brain For brain volume measurement (BVM), High-resolution (HR) MR images have shown to provide accurate results at small subcortical areas. However, prolonged scan time remains a classical challenge for 3D MRI. We implemented a combined technique, deep learning based super-resolution (DL-SR) and low-resolution Compressed Sensing (CS) 3D-TFE-T1W with acceleration factor 4 to generate Super-resolution (SR) images under one minute scan time. The results show that DL-SR model is able to improve image resolution, in which no significant differences (p>0.01) in quantitative volumetric measurement between reference and DL-SR at subcortical regions, except for caudate region. |
| 4188 | Computer 102
|
Denoising highly accelerated T2*-weighted brain MRI using a deep learning convolutional neural network |
| Bryan Quah1, Sreekanth Madhusoodhanan Nair1, Jin Jin2, Fei Han3, Brian Renner1, Elaina Gombos1, Ke Cheng Liu3, Sunil Patil3, John A. Derbyshire4, Ken Sakaie5, Emmanuel Obusez5, Jonathan Lee5, Mark Elliot6, Russell T. Shinohara7, Matthew K. Schindler8, Jae W. Song6, Michel Bilello6, Marwa Kaisey1, Nader Binesh9, Marcel Maya9, Javier Galvan9, Hui Han10, Debiao Li10, Andrew Solomon11, Daniel S. Reich12, Nancy L. Sicotte1, Mark Lowe5, Daniel Ontaneda13, Omar Al-Louzi1, and Pascal Sati1,10 | ||
1Department of Neurology, Cedars-Sinai Medical Center, Los Angeles, CA, United States, 2Siemens Healthcare Pty Ltd, Brisbane, Australia, 3Siemens Medical Solutions, PA, United States, 4Functional MRI Facility, National Institute of Mental Health, National Institutes of Health, Bethesda, MD, United States, 5Imaging Institute, Cleveland Clinic, Cleveland, OH, United States, 6Department of Radiology, University of Pennsylvania, Philadelphia, PA, United States, 7Department of Biostatistics, Epidemiology, and Informatics, Penn Statistics in Imaging and Visualization Center, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, United States, 8Department of Neurology, University of Pennsylvania, Philadelphia, PA, United States, 9Department of Imaging, Cedars-Sinai Medical Center, Los Angeles, CA, United States, 10Biomedical Imaging Research Institute, Cedars-Sinai Medical Center, Los Angeles, CA, United States, 11Larner College of Medicine, The University of Vermont, Burlington, VT, United States, 12Translational Neuroradiology Section, National Institute of Neurological Disorders and Stroke, National Institutes of Health, Bethesda, MD, United States, 13Mellen Center, Department of Neurology, Neurological Institute, Cleveland Clinic, Cleveland, OH, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Data Processing, Denoising, Neuroimaging The scan time of high-resolution T2*-weighted brain imaging using 3D echo-planar imaging (3D-EPI) can be significantly reduced by applying Controlled Aliasing In Parallel Imaging Results In Higher Acceleration (CAIPIRINHA). However, this comes at the expense of a significant reduction in image quality. In this study, we evaluated the feasibility of using a deep learning-based approach (DnCNN) to denoise highly accelerated 3D-EPI scans acquired at 3T. Our results show that DnCNN was able to efficiently denoise highly accelerated T2*-weighted brain scans while preserving anatomical and pathological details. |
| 4189 | Computer 103
|
Enhanced-Deep-Super-Resolution Neural Network on Multiple MR Brain Images |
| Cristiana Fiscone1, Nico Curti2, Matti Ceccarelli3, David Neil Manners1, Gastone Castellani2, Caterina Tonon1,4, Daniel Remondini5,6, Raffaele Lodi1,4, and Claudia Testa4,5 | ||
1Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy, 2Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, Bologna, Italy, 3Department of Agricultural and Food Sciences, University of Bologna, Bologna, Italy, 4Functional and Molecular Neuroimaging Unit, IRCCS Istituto delle Scienze Neurologiche di Bologna, Bologna, Italy, 5Department of Physics and Astronomy, University of Bologna, Bologna, Italy, 6INFN Bologna, Bologna, Italy |
||
Keywords: Machine Learning/Artificial Intelligence, Visualization, Super Resolution, Generalization Enhanced Deep Super Resolution (EDSR) is a machine learning model aimed to improve image spatial resolution. It was previously trained with general purpose figures and, in this work, directly tested on different MR images: T1w, T2w and Quantitative Susceptibility Mapping (QSM), a quantitative imaging technique. The studied cohort included 28 healthy subjects. Without needing fine-tuning, EDSR shows excellent ability of generalization over new kind of data, improving imaging visualization and outperforming the traditional bicubic upsampling. In future applications, images of patients will be considered to test EDSR reconstruction when there is pathological tissue. |
| 4190 | Computer 104
|
CNN-based three-dimensional superresolution technology in Brain MRI with generalized q-sampling imaging |
| Chun-Yuan Shin1, Yi-Ping Chao2, Li-Wei Kuo3,4, Yi-Peng Eve Chang5, and Jun-Cheng Weng1,6,7 | ||
1Department of Medical Imaging and Radiological Sciences, and Department of Artificial Intelligence, Chang Gung University, Taoyuan, Taiwan, 2Department of Computer Science and Information Engineering, and Graduate Institute of Artificial Intelligence, Chang Gung University, Taoyuan, Taiwan, 3Institute of Biomedical Engineering and Nanomedicine, National Health Research Institutes, Miaoli, Taiwan, 4Institute of Medical Device and Imaging, National Taiwan University College of Medicine, Taipei, Taiwan, 5Department of Counseling and Clinical Psychology, Columbia University, New York City, NY, United States, 6Medical Imaging Research Center, Institute for Radiological Research, Chang Gung University and Chang Gung Memorial Hospital at Linkou, Taoyuan, Taiwan, 7Department of Psychiatry, Chang Gung Memorial Hospital, Chiayi, Taiwan |
||
Keywords: Machine Learning/Artificial Intelligence, Diffusion/other diffusion imaging techniques Understanding neural connections helps scientists conduct cognitive behavioral research. There are many nerve fiber intersections in the brain that need to be observed, and the size is between 30-50 nanometers. Improving image resolution has become an important issue. Generalized q-sampling imaging (GQI) was used to reveal the fiber geometry of straight and crossing. However, it is difficult to accurately describe fiber bending, fanning, and diverging with low-resolution imaging. In this work, we tried to achieve superresolution with a deep learning method on diffusion magnetic resonance imaging (MRI) images that has the potential to assess crossing, curving, and splaying fiber structures. |
| 4191 | Computer 105
|
Artificial intelligence-based denoising for clinical magnetic resonance imaging: from head to toe |
| Dallas Turley1,2, Pattana Wangaryattawanich2, Jalal Andre2, Majid Chailan2, Johannes M. Peeters3, Kim Van de Ven3, and Orpheus Kolokythas2 | ||
1Philips Healthcare, Seattle, WA, United States, 2Department of Radiology, University of Washington, Seattle, WA, United States, 3Philips Healthcare, Best, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Artificial intelligence in MR is a diverse and growing field. In this work, we apply convolutional neural networks (CNN) to accelerated routine clinical protocols to investigate potential image quality improvements. For moderate increase in reconstruction time, CNNs were judged by experienced radiologists to significantly improve image quality by reducing noise and artifact. |
| 4192 | Computer 106
|
3D MRI super-resolution using convolutional generative adversarial network with gradient guidance |
| Wei Xu1, Jing Cheng1, and Dong Liang1 | ||
1Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, ShenZhen, China |
||
Keywords: Machine Learning/Artificial Intelligence, Brain The application of MRI has been limited due to the restriction of imaging time and spatial resolution. Super-resolution is an important strategy in clinics to speed up MR imaging. In this work, we propose a novel GAN-based super-resolution method which incorporates gradient features to improve the recovery of local structures of the super-resolution images. Experiments on 3D MR Vessel Wall imaging demonstrate the superior performance of the proposed method. |
| 4193 | Computer 107
|
Convolutional-neural-network-based denoising with estimated noise-based normalization to effectively reduce noise for various noise levels |
| Atsuro Suzuki1, Tomoki Amemiya1, Yukio Kaneko1, Suguru Yokosawa1, and Toru Shirai1 | ||
1Imaging Technology Center, FUJIFILM Corporation, Tokyo, Japan |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Noise reduction To develop a convolutional neural network (CNN)-based denoiser for various noise levels, we propose the use of estimated noise-based normalization in denoising. When the CNN-based denoiser with estimated noise-based normalization was applied to brain FLAIR images with various noise levels, it resulted in values closer to the normalized root mean square error (NRMSE) between the denoised and the target images compared with a conventional CNN-based denoiser trained with the same noise level as that in the input image. In conclusion, our method effectively reduced the noise in an image with various noise levels in terms of minimization of the NRMSE. |
| 4194 | Computer 108
|
Knowledge Distillation Enables Efficient Neural Network for Better Generalizability in MR Image Denoising and Super Resolution |
| Qinyang Shou1, Zechen Zhou2, Kevin Blansit2, Praveen Gulaka2, Enhao Gong2, Greg Zaharchuk2, and Ajit Shankaranarayanan2 | ||
1University of Southern California, Los Angeles, CA, United States, 2Subtle Medical Inc., Menlo Park, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Knowledge Distillation In this work, knowledge distillation (KD) is investigated to improve model generalizability for image enhancement tasks. KD can allow a 35× faster convolutional network to achieve similar performance as a Transformer based model in image denoising tasks. In addition, KD can enable a single image enhancement model for both denoising and super-resolution tasks that outperforms the conventional multi-task model trained with mixed data. KD potentially allows efficient image enhancement models to achieve better generalization performance for clinical translation. |
| 4195 | Computer 109
|
Clinical assessment of a deep learning-based denoising method for DWI of Nasopharyngeal Cancer |
| Tiebao Meng1, Haibin Liu1, Haoqiang He1, Jialu Zhang2, and Long Qian2 | ||
1Department of Radiology, Sun Yat-Sen University Cancer Center, Guangzhou, China, 2MR Research, GE Healthcare, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Diffusion/other diffusion imaging techniques Deep Learning reconstruction (DLR) has the potential to reduce MRI scan time while improving signal-to-noise ratio (SNR) and maintaining spatial resolution. This study evaluated results of DLR in 66 patients undergoing clinical DWI of nasopharyngeal cancer (NPC). To assess the image quality of the DWI with DLR, each patient underwent three different protocols: conventional DWI without DLR, fast DWI without DLR and fast DWI with DLR. The image quality was evaluated among these groups. Preliminary results suggested the feasibility of fast DWI with DLR in the diagnosis of NPC with reduced scan time. |
| 4196 | Computer 110
|
Improving SNR of high resolution multi-echo SWI using complex domain DL based denoising |
| Florintina C1, Sajith Rajamani1, Preetham Shankpal1, Suresh Joel1, Sudhanya Chatterjee1, Rohan Patil1, Ramesh Venkatesan1, Raja Sundaresan1, and Harsh Agarwal1 | ||
1GE Healthcare, Bengaluru, India |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence SWI is a high resolution MRI sequence, particularly sensitive to compounds which distort the local magnetic field making it useful in detecting blood products, calcium, etc. 3 and is used as part of brain MR imaging. The phase images are high pass filtered to remove the slow varying susceptibility changes and this is important to differentiate between para and diamagnetic substances. When this filtered phase image is used to accentuate the directly observed signal loss in the magnitude image, it is raised to a higher power and noise gets magnified as well, imparting undesirable effects in the SWI image. |
| 4197 | Computer 111
|
An MR-assisted Spatiotemporal Approach for 4D Dynamic Brain PET Denoising |
| Hamed Yousefi1, Chunwei Ying2, Mahdjoub Hamdi2, Richard Laforest2, and Hongyu An2 | ||
1Imaging Science, Washington University in St.Louis, St Louis, MO, United States, 2Washington University in St.Louis, St Louis, MO, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Brain In this study, dynamic brain PET denoising was done using registered MRI and reconstructed PET images reconstructed by the OSEM method, while maintaining TACs that were quite near to the original noisy data. The fundamental challenge to using the supervised learning approach is the lack of ground truth. The resulting spatiotemporal images improved the image quality according to various CNR, SNR, and CRC parameters while maintaining TACs that were similar to the original raw PET. This method only needs one trained network, one set of matrices for a statistical temporal PCA model, and 4D dynamic PET data as input. |
| 4198 | Computer 112
|
Improvement of canine brain FLAIR image using super-resolution GAN |
| Yuseoung Son1, Sumin Roh1, Jae-Kyun Ryu1, Won Beom Jung1, Seok-Mn Lee2, A-Rim Lee2, Chuluunbaatar Otgonbaatar3, Jaebin Lee4, Ho-Jung Choi2, Young-Won Lee2, and Hackjoon Shim1,4 | ||
1Medical Imaging AI Research Center, Canon Medical Systems Korea, Seoul, Korea, Republic of, 2College of Veterinary Medicine, Chungnam National University, Daejeon, Korea, Republic of, 3College of Medicine, Seoul National University, Seoul, Korea, Republic of, 4Magnetic Resonance Business Unit, Canon Medical Systems Korea, Seoul, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Most veterinary imaging has been achieved using human MRI scanners. Therefore, extensive averaging is required to obtain high-resolution images with high SNR for the animals, thereby leading to a long scan time. Veterinary MRI is typically performed under general anesthesia to minimize the level of stress and movement during image scanning. Therefore, long anesthetic conditions could affect animal normal physiology and be life-threatening, especially for patients in veterinary medical field. Here, we aimed to obtain higher image quality with short scanning time using super-resolution generative adversarial network (SRGAN) in the canine brain MRI. |
| 4199 | Computer 113
|
Coil2Coil 3D: Improved self-supervised MR image denoising using phased-array coil images based on 3D structure information |
| Rokgi Hong1, Juhyung Park1, Hayeon Lee2, Hongjun An1, and Jongho Lee1 | ||
1Department of Electrical and Computer Engineering, Seoul National University, Seoul, Korea, Republic of, 2School of Electrical Engineering, Korea University, Seoul, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence A deep-learning-based denoising method by extending Coil2Coil method based on 3D network was proposed. This method utilizes 3D information to conserve more tissue structure details. When tested for the synthetic noise-added images, the quantitative metrics were improved than using 2D network (PSNR: 46.57 ± 2.50 for 3D vs. 45.96 ± 2.47 for 2D). For the real-world noise, the proposed method also demonstrated qualitatively improved denoising than 2D case by conserving the structure details in MR images. Additionally, the quantitative mapping experiment for R2* showed the more accurate mapping results, suggesting improvement of the quantification by 3D denoising. |
| 4200 | Computer 114
|
Increased Resolution and Shortened Scan Time of T2 weighted Prostate MRI using Deep Learning Denoising Reconstruction at 1.5T |
| Hung Phi Do1, Dawn Berkeley1, Brian Tymkiw1, Wissam AlGhuraibawi1, and Mo Kadbi1 | ||
1Canon Medical Systems USA, Inc., Tustin, CA, United States |
||
Keywords: Prostate, Prostate In prostate MRI, high SNR images are desired for a better depiction of anatomical and pathological structures; however, it often requires a longer scan time, especially at 1.5T. Deep Learning Denoising Reconstruction (DLR) has been shown to effectively remove noise, allowing high SNR with higher resolution and shortened scan time simultaneously. This study demonstrates that DLR enables the acquisition of high SNR images with 29.48% scan time reduction and 31.51% increase in spatial resolution at 1.5T. |
| 4201 | Computer 115
|
Multi-Shot M1-Nulled Pancreatic Diffusion Weighted Imaging with Deep Learning-Based Denoising |
| Kang Wang1, Matthew J. Middione1, Andreas Markus Loening1, Patricia Lan2, Xinzeng Wang3, Daniel B Ennis1, and Ryan Lennex Brunsing1 | ||
1Radiology, Stanford University, Palo Alto, CA, United States, 2GE HealthCare, Menlo Park, CA, United States, 3GE HealthCare, Houston, TX, United States |
||
Keywords: Pancreas, Diffusion/other diffusion imaging techniques, Diffusion Weighted Imaging, Denoising, M1-nulled, ADC Multi-shot DWI (msDWI) may be improved by M1 motion-compensation, but at a penalty of longer TE and lower SNR. DL-based denoising has recently emerged as an option for DWI, which could help offset this TE penalty. Here we assess the impact of a commercially available DL-based denoising tool in the setting of motion-compensated msDWI. |
| 4202 | Computer 116
|
Deep Learning Denoising Reconstruction Enables up to 2X Scan Time Reduction while Maintaining Image Quality for Musculoskeletal Imaging |
| Hung Phi Do1, Dawn Berkeley1, Brian Tymkiw1, Wissam AlGhuraibawi1, and Mo Kadbi1 | ||
1Canon Medical Systems USA, Inc., Tustin, CA, United States |
||
Keywords: MSK, MSK Two-average (2NAQ) scans are often required to acquire high resolution images with adequate Signal-to-Noise Ratio (SNR). Deep Learning Denoising Reconstruction (DLR) effectively removes noise hence improving the SNR of reconstructed images. The SNR gain could be used to increase resolution and/or reduce scan time. This study demonstrates that the DLR-reconstructed one-average (1NAQ) images have similar image quality to those acquired with 2NAQ and reconstructed with conventional reconstruction. As a result, approximately 2X scan time reduction is achieved with DLR. |
| 4203 | Computer 117
|
Assessment of Multi-modal MR Imaging for Glioma Based on a Deep Learning Reconstruction Approach with the Denoising Method |
| Jun Sun1, Yiding Guo1, Zhizheng Zhuo1, Siyao Xu1, Min Guo1, Li Chai1, Junjie Li1, Liying Qu1, Minghao Wu1, Juan Wei2, Mingna Li3, Tong Li3, Jinyuan Weng4, Xiaodong Gong5, Yunyun Duan1, Dabiao Zhou1, and Yaou Liu1 | ||
1Capital Medical Universtiy, Beijing Tiantan Hospital, Beijing, China, 2MR Research, GE Healthcare, Beijing, China, 3BioMind Inc., Beijing, China, 4Department of Medical Imaging Product, Neusoft, Group Ltd., Shenyang, China, 5Department of Medical Imaging Product, Neusoft, Group Ltd., Beijing, China |
||
Keywords: Tumors, Brain Deep learning reconstruction (DLR) approach with denoising can improve image quality of magnetic resonance (MR) images. However, its applications on multi-modal glioma imaging have not been assessed.Multi-modal images of 107 glioma patients were evaluated by signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), edge sharpness, visual assessment, diagnosis accuracy and efficiency. Contrasted with conventionally reconstructed images, the DLR images showed higher tumor/residual tumor SNR, higher tumor to white/gray matter CNR, better results of the visual assessment, and a trend of improved diagnosis efficiency and comparable accuracy. DLR can improve image quality of multi-modal glioma images which should benefit the glioma diagnosis. |
| 4204 | Computer 118
|
Fast MRI of the cervical spinal cord by combining the deep learning-based denoising and Super-Resolution technique |
| Takuya Aoike1, Noriyuki Fujima2, Jihun Kwon3, Masami Yoneyama3, Satonori Tsuneta2, Kinya Ishizaka1, and Kohsuke Kudo4 | ||
1Department of Radiological Technology, Hokkaido University Hospital, Sapporo, Japan, 2Department of Diagnostic and Interventional Radiology, Hokkaido University Hospital, Sapporo, Japan, 3Philips Japan, Ltd., Tokyo, Japan, 4Department of Diagnostic Imaging, Hokkaido University Graduate School of Medicine, Sapporo, Japan; Global Center for Biomedical Science and Engineering, Faculty of Medicine, Hokkaido University, Sapporo, Japan |
||
Keywords: Spinal Cord, Spinal Cord SmartSpeed Precise Image (SSPI) is one of novel image reconstruction technique which combined the physics driven type deep learning-based denoising process and Super-Resolution techniques. We investigated the utility of SSPI-based single-shot turbo spin-echo sequence of the cervical spinal cord T2-weighted image with the short acquisition time. High quantitative signal-to-noise ratio and qualitative image sharpness were successfully demonstrated by using SSPI. This methodology can be useful for the better image quality and the short acquisition time in the assessment of patients with cervical spinal cord diseases. |
| 4205 | Computer 119
|
Patient-specific Self-supervised Resolution-enhancing Networks for Synthesizing High-resolution Magnetic Resonance Images |
| Xiaofeng Yang1, Sagar Mandava2, Yang Lei1, Huiqiao Xie1, Tonghe Wang3, Justin Roper1, Tian Liu4, and Hui Mao1 | ||
1Emory University, Atlanta, GA, United States, 2GE Healthcare, Atlanta, GA, United States, 3Memorial Sloan Kettering Cancer Center, New York, NY, United States, 4Icahn School of Medicine at Mount Sinai, New York, NY, United States |
||
Keywords: Quantitative Imaging, Data Processing This study aims to develop an efficient and clinically applicable method using patient-specific self-supervised resolution-enhancing network to synthesize the high-resolution information of MR images in the low-resolution direction to generate respective high-resolution MRI. |
| 4342 | Computer 81
|
Simultaneous Liver and Spleen Segmentation using U-Net Transformer Model on T1-weighted and T2-weighted MRI Data |
| Huixian Zhang1, Redha Ali1, Hailong Li1, Wen Pan1, Scott B. Reeder2, David T. Harris2, William R. Masch3, Anum Alsam3, Krishna P. Shanbhogue4, Nehal A. Parikh1, Jonathan R. Dillman1, and Lili He1 | ||
1Cincinnati Children's Hospital, Cincinnati, OH, United States, 2University of Wisconsin-Madison, Madison, WI, United States, 3Michigan Medicine, University of Michigan, Ann Arbor, MI, United States, 4NYU Langone Health, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence To develop an AI system for precise and fully automated, simultaneous segmentation of the liver and spleen from T1-weighted and T2-weighted MRI data. Our study compares the performance of the U-Net Transformer (UNETR) and the standard 3D U-Net model for simultaneous liver and spleen segmentation using MRI images from pediatric and adult patients from multiple institutions. Our work demonstrates that the UNETR shows a statistical improvement over 3D U-Net in both liver and spleen segmentations. |
| 4343 | Computer 82
|
A Deep Learning-based Multi-Contrast MRI Registration Model with a Realistic Flow Field and Reduced Over-Smoothing Effect |
| Yiheng Li1 and Ryan Chamberlain1 | ||
1Subtle Medical Inc., Menlo Park, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Data Processing, Registration To optimize DL-based medical registration models to produce a more realistic flow field and reduce the over-smoothing effect of deformable registration while keeping the generalizability of the multi-contrast registration model to work on all anatomical structures and contrast of MRI, we proposed an algorithm that adopts the image synthetic framework from SynthMorph and optimized with cycle-consistent loss. By experimenting with the jacobian loss, bidirectional loss, and cycle-consistent loss, we managed to further optimize the results of registered images. The evaluation of two MRI image datasets, the BraTS dataset, and the LSpine dataset, demonstrated the increased SSIM, PSNR, and LocalNCC. |
| 4344 | Computer 83
|
Spatial-Adaptive Deep Learning Model and Magnetic Resonance Fingerprinting for Segmentation and Quantitative Evaluation of Cervical Cancer |
| Reza Kalantar1, Jessica Mary Winfield1,2, Mihaela Rata1, Gigin Lin3, Susan Lalondrelle1,4, Christina Messiou1,2, Matthew David Blackledge1, and Dow-Mu Koh1,2 | ||
1Radiotherapy and Imaging, The Institute of Cancer Research, London, United Kingdom, 2Radiology, The Royal Marsden Hospital, London, United Kingdom, 3Medical Imaging and Intervention, Chang Gung Memorial Hospital at Linkou and Chang Gung University, Taoyuan, Taiwan, 4Gynaecological Unit, The Royal Marsden Hospital, London, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Cancer Quantitative magnetic resonance imaging (qMRI) can provide additional information for diagnosis and response assessment, but adoption of multi-parametric qMRI techniques has been hindered by long acquisition times and labor-intensive processing steps. Magnetic resonance fingerprinting (MRF) provides quantitative maps in a single acquisition but MRF deployment in clinical studies still requires manual delineation of volumes of interest. A spatial-adaptive deep learning framework was developed to segment cervical cancer on MRI and quantify T1 relaxation times of the tumor pre- and post-radiotherapy treatment. Our results suggest that automated segmentation models may be promising tools for quantitative tumor evaluation and treatment response assessment. |
| 4345 | Computer 84
|
ZTE segmentation of glenohumeral bone structure using deep learning |
| Michael Carl1, Kaustaub Lall2, Armin Jamshidi2, Eric Chang3,4, Sheronda Statum3,4, Anja Brau1, Christine B Chung3,4, Maggie Fung1, and Won C. Bae3,4 | ||
1General Electric Healthcare, Menlo Park, CA, United States, 2Electrical and Computer Engineering, University of California, San Diego, La Jolla, CA, United States, 3Radiology, University of California, San Diego, La Jolla, CA, United States, 4Radiology, VA San Diego Healthcare System, San Diego, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation, ZTE, Deep learning Evaluation of 3D bone morphology of the glenohumeral joint is necessary for pre-surgical planning. Zero echo time (ZTE) MRI provides excellent bone contrast, and we developed a deep learning model to perform automated segmentation of major bones (i.e., humerus and others) from ZTE to aid evaluation. Axial ZTE images of normal shoulders (n=31) acquired at 3T were annotated for training with a 2D U-Net, and the trained model was validated with testing data (n=10 normal shoulder, n=6 symptomatic). Testing accuracy was around 80 to 90% (Dice score) for either cohort, except for a few failed cases with very low scores. |
| 4346 | Computer 85
|
Novel Uunsupervised Segmentation of Bone Marrow Edema-Like Lesions using Bayesian Conditional Generative Adversarial Networks |
| Andrew Seohwan Yu1,2,3, Sibaji Gaj1,2, William Holden1,2, Richard Lartey1,2, Jeehun Kim1,2,4, Carl Winalski1,2,5, Naveen Subhas1,2,5, and Xiaojuan Li1,2,5 | ||
1Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, OH, United States, 2Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, United States, 3Cleveland Clinic, Department of Computer and Data Sciences, Case Western Reserve University, Cleveland, OH, United States, 4Department of Electrical, Computer, and Systems Engineering, Case Western Reserve University, Cleveland, OH, United States, 5Department of Diagnostic Radiology, Imaging Institute, Cleveland Clinic, Cleveland, OH, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation Quantitative assessment of the bone marrow edema-like lesions and their association with osteoarthritis requires a consistent and unbiased segmentation method, which is difficult to obtain in the presence of human annotators. This study proposes an unsupervised approach using Bayesian deep learning and conditional generative adversarial networks that detects and segments anomalies without human intervention. The full pipeline has a lesion-wide sensitivity of 0.86 on unseen scans. This approach is expected to be generalizable to other lesions and/or modalities. |
| 4347 | Computer 86
|
3D nnU-Net with Multi Loss Ensembles for Automated Segmentation of Intracranial Aneurysm |
| Maysam Orouskhani1, Shaojun Xia2, Mahmud Mossa-Basha1, and Chengcheng Zhu1 | ||
1Department of Radiology, University of Washington, Seattle, WA, United States, 2Peking University Cancer Hospitals & Institution, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Compound Loss function, Deep Neural Networks, nnU-Net In the segmentation of intracranial aneurysm, deep neural networks are equipped with modified loss functions to penalize the training weights for aneurysm false predictions and conduct unbiased learning. In this paper, we used a new compound loss function to capture the different aspects of embedding as well as diverse features. The proposed loss was given to a 3D full resolution nnU-Net to segment imbalanced TOF-MRA images from ADAM dataset. The proposed loss outperformed commonly used losses in terms of Dice, Sensitivity, and Precision. |
| 4348 | Computer 87
|
Cerebrovascular segmentation of mouse TOF-MRA using 3D U-Net |
| Yue Wu1,2, Jinyuan Zhang3,4, Zhixin Li3,4, Yishuang Yang3,4, Yan Zhuo3,4, Xudong Zhao2,3, and Zihao Zhang2,3 | ||
1AHU-IAI AI Joint Laboratory, Anhui University, Hefei, China, 2Institute of Artificial Intelligence, Hefei Comprehensive National Science Center, Hefei, China, 3State Key Laboratory of Brain and Cognitive Science, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China, 4University of Chinese Academy of Sciences, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Vessels Time-of-flight MR angiography (TOF-MRA) allows noninvasive visualization of the mouse intracranial vessels. However, there is no available tool for vessel segmentation in mouse 3D TOF-MRA. In this study, we developed an automated vessel segmentation method based on a convolutional neural network and multiscale vessel enhancement filtering algorithm. The method is potentially useful in the preclinical studies of cerebrovascular diseases. |
| 4349 | Computer 88
|
Automated deep learning based segmentation of abdominal adipose tissue on whole-body MRI in a population-based study of adolescents |
| Tong Wu1, Santiago Estrada2, Renza Gils1, Ruisheng Su1, Vincent Jaddoe3, Edwin Oei1, and Stefan Klein1 | ||
1Department of Radiology and Nuclear Medicine, Erasmus MC, Rotterdam, Netherlands, 2Image Analysis, German Center for Neurodegenerative Diseases (DZNE), Bonn, Germany, 3Department of Pediatrics, Erasmus MC, Rotterdam, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation, Dixon MRI This study is embedded in the Generation R Study, a population-based prospective cohort study in the Netherlands. Whole-body Dixon MRI scans were performed at age 13 years. A previous neural network (CDFNet) has been published that was trained on adults. We aimed to retrain this network on our MRIs in the Generation R Study. The obtained segmentations are in strong agreement with expert-generated manual segmentations and can therefore greatly reduce the manual workload. Therefore, for accurate abdominal fat quantification, segmentation of both subcutaneous and visceral adipose tissue, on Dixon MRIs of adolescents is feasible with a retrained convolutional neural network. |
| 4350 | Computer 89
|
3D segmentation of MRI images of leg muscles using Convolutional Neural Networks |
| Lara Schlaffke1, Niklas Doliwa1, Marius Markmann1, Marlena Rohm1,2, Robert Rehmann1,3, and Martijn Froeling4 | ||
1Neurology, University Clinic Bergmannsheil Bochum gGmbH, Bochum, Germany, 2Heimer Institute for Muscle Research, University Clinic Bergmannsheil Bochum gGmbH, Bochum, Germany, 3Neurology, Klinikum Dortmund, University Witten-Herdecke, Dortmund, Germany, 4Radiology, UMC Utrecht, Utrecht, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Muscle, Segmentation Quantitative magnetic resonance imaging offers promising surrogate biomarkers for the early diagnosis and monitoring of pathological changes in leg muscles in patients with neuromuscular disorders. To use this in a clinical routine, automatic segmentation of muscles is needed. To investigate precision and accuracy of various Computational neuronal networks for automated segmentation of upper leg muscles, different loss function were used and quantitative outcome compared to the gold-standard of manual segmentation. The best segmentation results were achieved, with the Fokal Tversky loss function and a single input approach. |
| 4351 | Computer 90
|
A deep learning-based liver tumor segmentation algorithm in enhanced multi–phase MRI images |
| Meng Dou1,2, Ying Zhao 3, Tao Lin3, Yu Yao1,2, and Ailian Liu3 | ||
1Chengdu institute of computer application, Chinese academy of sciences, Chengdu, China, 2University of Chinese Academy of Sciences, Beijing, China, 3Department of Radiology, First Affiliated Hospital of Dalian Medical University, Dalian, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence In this paper, we propose a deep learning-based liver tumor segmentation algorithm in enhanced multi-phase MRI images. The experimental results show that the proposed method can segment liver and tumor in enhanced multi-phase MRI images with a smaller resource occupation, and outperforms the comparison method. |
| 4352 | Computer 91
|
A Non-linear, Tissue-specific 3D Data Generative Framework for Deep Learning Segmentation Performance Enhancement |
| Soumya Ghose1, Chitresh Bhushan1, Dattesh Shanbhag2, Desmond Teck Beng Yeo3, and Thomas K. Foo3 | ||
1AI and Computer Vision, GE Research, Niskayuna, NY, United States, 2GE Healthcare, Bangalore, India, 3Biology & Physics, GE Research, Niskayuna, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation, Vertebra segmentation, T2w, Deep Learning, Intensity Transformation Deep learning (DL) models have been successful in solving segmentation problems that involve large, balanced, and labeled datasets. However, in the medical imaging domain, it is rare to find manually annotated datasets that capture the entire spectrum of heterogeneity. Novel datasets with significantly different intensities than training datasets, may adversely affect DL model performance. In this work, we present a hybrid framework for tissue-specific, non-linear intensity transformation of pediatric T2w images similar to that of the adults training dataset and demonstrate an improved performance for vertebra segmentation of pediatric datasets without the need for DL network re-training/re-tuning. |
| 4353 | Computer 92
|
A novel federated learning framework for accurate and secure multi-center MS lesion segmentation |
| Dongnan Liu1,2, Mariano Cabezas2, Dongang Wang2,3, Zihao Tang1,2, Geng Zhan2,3, Kain Kyle2,3, Linda Ly2,3, James Yu2,3, Chun-Chien Shieh2,3, Ryan Sullivan4, Fernando Calamante4,5, Michael Barnett2,3, Wanli Ouyang6, Weidong Cai1, and Chenyu Wang2,3 | ||
1the School of Computer Science, University of Sydney, Sydney, Australia, 2Brain and Mind Centre, University of Sydney, Sydney, Australia, 3Sydney Neuroimaging Analysis Centre, Sydney, Australia, 4the School of Biomedical Engineering, University of Sydney, Sydney, Australia, 5Sydney Imaging, University of Sydney, Sydney, Australia, 6the School of Electrical and Information Engineering, University of Sydney, Sydney, Australia |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation Multiple sclerosis (MS) is a neurodegenerative disease of the central nerve system (CNS), which has the potential to cause a neurological disability, particularly for young adults. Recently, deep learning-based techniques are important for MS diagnosis and treatment, since they can segment the lesions caused by MS automatically and accurately. However, their applicability to multi-center scenarios is limited, due to the privacy and security issues in data sharing. To tackle these limitations, a decentralized deep learning framework is designed in this work, which can bring accurate multi-center MS lesion segmentation performance without sharing the raw data. |
| 4354 | Computer 93
|
Accelerated cardiac cine imaging using deep learning: impact on left ventricular function and AI segmentation model |
| Wenyun Liu1, Chinting Wong2, Cheng Li3, Di Yu1, Yi Zhu4, Ke Jiang4, and Huimao Zhang1 | ||
1Radiology, The First Hospital of Jilin University, Changchun, China, 2Nuclear Medicine, The First Hospital of Jilin University, Changchun, China, 3Cardiovascular Center, The First Hospital of Jilin University, Changchun, China, 4Philips Healthcare, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Myocardium, Cardiomyopathy,Data Analysis Cardiac MRI is considered reference standard for the noninvasive assessment of ventricular volumes and function. However, long multibreath-hold acquisition time can prove difficult in patients and lead to poor image quality. In this study, we investigated the use of a deep learning-based reconstruction algorithm, named Compressed SENSE Artificial Intelligence(CS-AI), to accelerate two-dimensional cine bSSFP for cardiac MRI. The purpose of this study was to compare the image quality and performance of a CS-AI-based cine sequence between reference and accelerated methods: SENSE, Compressed-SENSE, and CS-AI, and then to investigate the impact of images reconstructed by deep learning on AI segmentation model. |
| 4355 | Computer 94
|
Amniotic Fluid Segmentation using Convolutional Neural Networks |
| Alejo Costanzo1,2, Birgit Ertl-Wagner3,4, and Dafna Sussman1,5,6 | ||
1Department of Electrical, Computer and Biomedical Engineering, Toronto Metropolitan University, Toronto, ON, Canada, 2Institute for Biomedical Engineering, Science and Technology, Toronto Metropolitan University and St. Michael’s Hospital, Toronto, ON, Canada, 3Division of Neuroradiology, The Hospital for Sick Children, Toronto, ON, Canada, 4Department of Medical Imaging, Faculty of Medicine, University of Toronto, Toronto, ON, Canada, 5Institute for Biomedical Engineering, Science and Technology, Toronto Metropolitan University and St. Michael’s Hospital, Toronto, ON, Canada, 6Department of Obstetrics and Gynecology, Faculty of Medicine, University of Toronto, Toronto, ON, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation, Convolutional Neural Network Amniotic Fluid Volume (AFV) is an important fetal biomarker when diagnosing certain fetal abnormalities. We aim to implement a novel Convolutional Neural Network (CNN) model for amniotic fluid (AF) segmentation which can facilitate clinical AFV evaluation. The model, called AFNet was trained and tested on a radiologist–validated AF dataset. AFNet improves upon ResUNet++ through the efficient feature mapping in the attention block, and transpose convolutions in the decoder. Experimental results show that our AFNet model achieved a 93.38% mean Intersection over Union (mIoU) on our dataset. We further demonstrate that AFNet outperforms state-of-the-art models while maintaining a low model size. |
| 4356 | Computer 95
|
Brain tumor segmentation using out-of-distribution feature from normal MR images |
| Namho Jeong1, Beomgu Kang1, and Hyunwook Park1 | ||
1Korea Advanced Institute of Science and Technology, Daejeon, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Tumor Abundant data of healthy subjects is available and thus it would be helpful if normal data can boost the brain tumor segmentation. We proposed the out-of-distribution (OOD) feature based on the learned distribution of normal data to discriminate abnormal tumors. Once the neural network was trained to synthesize other contrast MR images with normal data, it failed to properly synthesize the tumor tissues. This OOD characteristic was used for the generation of new feature that can help the tumor segmentation. It was verified that the OOD features contributed to an improvement of segmentation through experiments. |
| 4357 | Computer 96
|
Breast Tumor Segmentation using U-Net with ResNet34 |
| Yunkyoung Jun1, Jiwoo Jeong1, Seokha Jin1, Noehyun Myung1, Jimin Lee2, and Hyungjoon Cho1 | ||
1BME, UNIST, Ulsan, Korea, Republic of, 2Nuclear Engineering, UNIST, Ulsan, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Animals The proposed research implemented an automatic tumor segmentation application on an orthotopic breast tumor model. This application can segment the tumors accurately and monitor tumor growth and the therapeutic effect of Doxorubicin for treatment. Also, the outputs from the application can reconstruct into 3D rendering and offer the visualization of shape and volume. As a result, the application can be applied to orthotopic breast tumor model research. |
| 4358 | Computer 97
|
Comparison of Combinations of MTR, FLAIR, and T1WCE MRI Images for Automatic Segmentation of Brain Metastases |
| Ananya Anand1, Daniel Nussey2, and Dafna Sussman3 | ||
1Toronto Metropolitan University, Fremont, CA, United States, 2Toronto Metropolitan University, Toronto, ON, Canada, 3Electrical, Computer, and Biomedical Engineering, Toronto Metropolitan University, Toronto, ON, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Magnetization transfer, Multi-Modal This study presents an evaluation of utilizing MTR and FLAIR as compared to T1wCE and FLAIR images for automatic segmentation of brain metastases, due to the unknown consequences posed by accumulated contrast enhancement. Numerous combinations of FLAIR, T1wCE, and MTR MRI images were tested on three top-performing segmentation models (MS-FCN, U-NET, and SegresNet) to evaluate the feasibility of using MTR and FLAIR images for tumor segmentation. Overall, the U-Net produced the best similarity scores of 0.4±0.15 using MTR and FLAIR images. A future study will test the utility of 3D MTR images for tumor segmentation using convolutional and full-connected models. |
| 4359 | Computer 98
|
Cycle Inverse Consistent Deformable Medical Image Registration with Transfomer |
| Chenghao Zhang1, Mikhail Morgan1, Yanting Yang1, Ye Tian1, and Jia Guo1 | ||
1Columbia University, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Brain Deformable image registration is a crucial part of the medical image process that seeks to estimate an optimal spatial transformation to align two images. Traditional methods neglect the inverse-consistent property and topology preservation of the transformation, which can lead to errors in registration results. To address these issues, we propose the cycle inverse consistent transformer-based deformable medical image registration model. We adopt Swin-UNet to achieve higher registration performance and comprehensively consider the inverse consistency loss function to guarantee a more accurate registration. Our pipeline can be trained to create study-specific templates of images for diagnostic and/or educational purposes. |
| 4360 | Computer 99
|
Deep learning Based Automatic Segmentation of Brain Magnetic Resonance angiography Images |
| Liying An1, Fanyang Meng1, Kaixin Yu2, Lei Zhang1, and Huimao Zhang1 | ||
1Radiology, The First Hospital of Jilin University, Changchun, China, 2Shukun (Beijing) Technology Co., Ltd, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Vessels Accurate cerebrovascular segmentation plays an important role in clinical diagnosis and the related research. Magnetic Resonance angiography (MRA) postprocessing manually recognized by technologists is extremely labor intensive and error prone. However, automatic segmentation of brain vessels remains challenging because of the variable vessel shape and high complex of vessel geometry. We propose an artificial intelligence brain vessel segmentation system supported by an optimized physiological anatomical-based 3D convolutional neural network that can automatically achieve brain MRA vessel segmentaion in healthcare services. The overall segmentation accuracy of the independent testing dataset is 0.931. |
| 4361 | Computer 100
|
Deep learning-based groupwise registration for longitudinal MRI analysis in glioma |
| Claudia Chinea Hammecher1,2, Karin van Garderen1,3, Marion Smits1,3, Pieter Wesseling4,5, Bart Westerman6, Pim French7, Mathilde Kouwenhoven8, Roel Verhaak9,10, Frans Vos1,2, Esther Bron1, and Bo Li1 | ||
1Department of Radiology & Nuclear Medicine, Erasmus MC, Rotterdam, Netherlands, 2Department of Imaging Physics, Delft University of Technology, Delft, Netherlands, 3Medical Delta, Delft, Netherlands, 4Department of Pathology, Amsterdam UMC / VU Medical Center, Amsterdam, Netherlands, 5Laboratory for Childhood Cancer Pathology, Princess Máxima Center for Pediatric Oncology, Utrecht, Netherlands, 6Department of Neurosurgery, Amsterdam UMC / VU Medical Center, Amsterdam, Netherlands, 7Department of Neurology, Erasmus MC Cancer Institute, Rotterdam, Netherlands, 8Department of Neurology, Amsterdam UMC / VU Medical Center, Amsterdam, Netherlands, 9The Jackson Laboratory for Genomic Medicine, Farmington, CT, United States, 10Department of Neurosurgery, Amsterdam UMC/VUmc, Amsterdam, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Data Analysis, Image Registration Glioma growth may be quantified with longitudinal image registration. However, the large mass-effects and tissue changes across images suposse an added challenge. Here, we propose a longitudinal, learning-based and groupwise registration method for the accurate and unbiased registration of glioma MRI. We evaluate on a dataset from the Glioma Longitudinal AnalySiS consortium and compare to classical registration methods. We achieve comparable Dice coeffients, with more detailed registrations, while significantly reducing the runtime to under a minute. The proposed methods may serve as an alternative to classical toolboxes, to provide further insight into glioma growth . |
| 4511 | Computer 81
|
The Yale Glioma Dataset: Developing An Open Access, Annotated MRI Database |
| Matthew L. Sala1, Jan Lost2, Niklas Tillmanns2, Sara Merkaj3, Marc von Reppert4, Divya Ramakrishnan1, Khaled Bousabarah5, Anita Huttner6, Sanjay Aneja7, Arman Avesta7, Antonio Omuro8, and Mariam Aboian1 | ||
1Radiology and Biomedical Imaging, Yale School of Medicine, New Haven, CT, United States, 2University of Düsseldorf, Düsseldorf, Germany, 3University of Ulm, Ulm, Germany, 4Leipzig University, Leipzig, Germany, 5Visage Imaging, Düsseldorf, Germany, 6Pathology, Yale School of Medicine, New Haven, CT, United States, 7Therapeutic Radiology, Yale School of Medicine, New Haven, CT, United States, 8Neurology, Yale School of Medicine, New Haven, CT, United States |
||
Keywords: Tumors, Machine Learning/Artificial Intelligence, Glioma Recent development of Machine Learning (ML) tools for analysis of CNS tumors demonstrates great potential benefit to research and clinical practice but has been hindered by a lack of external validation. There is a critical need for open access to large individual hospital-based datasets with expert annotations. Here, we present the Yale Glioma Dataset, a database of 1,033 patients featuring annotated segmentations on FLAIR and T1 post-gadolinium, tumor grading and classification, and further clinical information. Open access of this database will support the development and validation of new AI algorithms for glioma detection and segmentation. |
| 4512 | Computer 82
|
Efficient Fetal Brain Segmentation according to the Point Spread Function of MRI |
| Yunzhi Xu1, Jiaxin Li1, Xue Feng2, Kun Qing3, Dan Wu1, and Li Zhao1 | ||
1Zhejiang University, Hangzhou, China, 2Biomedical Engineering, University of Virginia, Virginia, VA, United States, 3Department of Radiation Oncology, City of Hope National Center, Los Angeles, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Fetus, Fetal Brain MRI，Point Spread Function High apparent resolution of fetal MRIs is provided by slice-to-volume reconstruction pipelines widely. However, the physical resolution of the fetal brain is lower than that. Therefore, we hypothesize that fetal brain segmentation can be performed based on downsampled fetal brain MRI according to its point spread function. In this work, 150 adult brain and 80 fetal brain MRIs were used to validate hypothesize. Using downsampled fetal data with factor of 4, a highly efficient segmentation model achieved similar segmentation accuracies compared to original data, which demonstrated that segmentation models can be developed based on PSF.
|
| 4513 | Computer 83
|
Impact of Label-Set on the Performance of Deep Learning-Based Segmentation of the Prostate Gland and Zones on T2-Weighted MR Images |
| Jakob Meglič1,2, Mohammed R. S. Sunoqrot1,3, Tone F. Bathen1,3, and Mattijs Elschot1,3 | ||
1Department of Circulation and Medical Imaging, Norwegian University of Science and Technology (NTNU), Trondheim, Norway, 2Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia, 3Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway |
||
Keywords: Machine Learning/Artificial Intelligence, Prostate Prostate segmentation is an essential step in computer-aided diagnosis systems for prostate cancer. Deep learning (DL)-based methods provide good performance for prostate segmentation, but little is known about the impact of ground truth (manual segmentation) selection. In this work, we investigated these effects and concluded that selecting different segmentation labels for the prostate gland and zones has a measurable impact on the segmentation model performance. |
| 4514 | Computer 84
|
Semi-supervised segmentation for 3D medical image based on contrast learning |
| Zhengyong Huang1,2, Na Zhang1, Dong Liang1, Xin Liu1, Hairong Zheng1, and Zhanli Hu1 | ||
1Lauterbur Research Center for Biomedical Imaging, Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, 2University of Chinese Academy of Sciences, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Data Processing, MRI medical segmentation Semi-supervised segmentation, using large amounts of unlabeled data and small amounts of labeled data, has achieved great success. This paper proposes a semi-supervised segmentation method based on consistent learning and contrast learning. It mainly uses a mean-teacher framework to add consistency losses and contrast losses based on multiscale features to minimize the distance of model responses under different disturbance inputs. In addition, mean square error loss was used to alternately minimize the gap between the teacher and student models. In 3D left atrium data, a Dice coeffivient of 0.8970 was obtained, which was superior to other methods. |
| 4515 | Computer 85
|
Self-Supervised Learning for Perivascular Spaces segmentation with enhanced contrast knowledge |
| Haoyu Lan1, Arthur W. Toga1, and Jeiran Choupan1,2 | ||
1University of Southern California, Los Angeles, CA, United States, 2NeuroScope Inc., Scarsdale, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation, self-supervised learning Due to the absence of isotropic T2w modality in clinical datasets, it is challenging to enhance the PVS contrast using multiple neuroimage modalities. To overcome this issue, in this work we introduced using self-supervised pre-trained model in the enhanced PVS contrast image space to improve the downstream model segmentation performance when solely using T1w as the training data. The experiment results showed that the proposed method increased segmentation accuracy compared to the model trained from scratch using T1w modality and resulted in faster training and less required training data volume. |
| 4516 | Computer 86
|
Improved segmentation of neuromelanin region for low SNR short scan sandwichNM imaging |
| Joonhyeok Yoon1, Juhyung Park1, Yoonho Nam2, Chul-Ho Sohn3,4, Junghwa Kang2, Jonghyo Youn1, Sooyeon Ji1, Chungseok Oh1, and Jongho Lee1 | ||
1Department of Electrical and Computer Engineering, Seoul National University, Seoul, Korea, Republic of, 2Department of Biomedical Engineering, Hankuk University of Foreign Studies, Gyeonggi-do, Korea, Republic of, 3Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, Korea, Republic of, 4Department of Radiology, Seoul National University Hospital, Seoul, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Parkinson's Disease, Neuromelanin, Denoising Neuromelanin (NM) has been considered an associated biomarker of Parkinson’s disease (PD). Conventional NM visualizeing techniques requires about 5~10min which is sub-optimal for scanning PD patient with movement disorders. Recently, SandwichNM is reduced scan time to 5m 30s, but it is still may not enough. In this research, we approach this issue in the viewpoint of image post processing with denoising techniques to reduce scan time. After 162 NM MRI data acquisition with short scan time, we demonstrated that using denoising technique can improve the distinguishability for NM segmentation. |
| 4517 | Computer 87
|
Improved Robustness for Deep Learning-based Segmentation of Perfusion CMR Using Data Adaptive Uncertainty-guided Spatiotemporal Analysis |
| Dilek Mirgun Yalcinkaya1,2, Khalid Youssef3, Bobak Heydari4, Subha Raman3,5, Rohan Dharmakumar3,5, and Behzad Sharif1,3,5 | ||
1Laboratory for Translational Imaging of Microcirculation, Indiana University (IU) School of Medicine, Indianapolis, IN, United States, 2Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, United States, 3Krannert Cardiovascular Research Center, IU School of Medicine/IU Health Cardiovascular Institute, Indianapolis, IN, United States, 4Stephenson Cardiac Imaging Centre, Department of Cardiac Sciences, University of Calgary, Calgary, AB, Canada, 5Weldon School of Biomedical Engineering, Purdue University, West Lafayette, IN, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation We proposed and validated Data Adaptive Uncertainty-Guided Spatiotemporal (DAUGS) analysis that leverages the data-driven uncertainty map of the segmentation contours among a pool of trained deep neural networks (DNNs) and automatically selects the segmentation result with the highest level of certainty. Our results suggest that proposed DAUGS and standard DNN-based analysis demonstrated on-par performance on the internal test set which is from the same institution as training set and acquired with FLASH sequence. In contrast, DAUGS analysis considerably outperformed DNN-based analysis on the external test set which was acquired with a bSSFP pulse sequence at a different institution, demonstrating the improved robustness of the proposed method despite limited training data. |
| 4518 | Computer 88
|
AI algorithms for classification of the mpMRI image sequences and segmentation of the prostate gland: an external validation study |
| Kexin Wang1 | ||
1Capital Medical University, Beijing, China |
||
Keywords: Prostate, Machine Learning/Artificial Intelligence In this study we evaluate the generalization of the AI algorithms for the classification of the mpMRI image sequences and the segmentation of the prostate gland with multicenter external dataset. A total of 719 patients who underwent multiparametric MRI (mpMRI) of the prostate were collected retrospectively from two hospitals. AI algorithms were tested for classification of the image type and segmentation of the prostate gland. The AI models demonstrated good performance in the external validation in the task of image classification and prostate gland segmentation. |
| 4519 | Computer 89
|
Reduced input combination study for the Simultaneous Multi-Tissue Segmentation and Multi-Parameter Quantification Network (MSMQ-Net) of Knee |
| Xing Lu1, Yajun Ma1, Jiyo Athertya1, Chun-Nan Hsu2, Eric Y Chang1,3, Amilcare Gentili1,3, Christine Chung1, and Jiang Du1 | ||
1Department of Radiology, University of California, San Diego, San Diego, CA, United States, 2Department of Neurosciences, University of California, San Diego, San Diego, CA, United States, 3Radiology Service, Veterans Affairs San Diego Healthcare System, San Diego, CA, United States |
||
Keywords: Osteoarthritis, Quantitative Imaging To accelerate ultrashort echo time (UTE) based multi-parameter quantitative MRI (qMRI), we performed comparison studies on the input MRI image numbers and the effects on the prediction quality of the MSMQ-Net. The MSMQ-net was modified and trained accordingly with different combinations of inputs. Both image similarity and regional analysis were evaluated. The results demonstrate that 90% accuracy can be achieved for both UTE-T1 and UTE-T1rho mapping when the scan time is reduced by 75%. |
| 4520 | Computer 90
|
Full and weak supervision networks for meniscus segmentation and multi classification based on MRI: data from the Osteoarthritis Initiative |
| Kexin Jiang1, Yuhan Xie2, Zhiyong Zhang2, Jiaping Hu1, Shaolong Chen2, Zhongping Zhang3, Changzhen Qiu2, and Xiaodong Zhang1 | ||
1Department of Medical Imaging, The Third Affiliated Hospital of Southern Medical University, GuangZhou, China, 2Electronics and Communication Engineering, Sun Yat-sen University, GuangZhou, China, 3Philips Healthcare, GuangZhou, China |
||
Keywords: Joints, Joints, meniscus Quantitative MRI of meniscus morphology (such as MOAKS system) have shown clinical relevance in the diagnosis of osteoarthritis. However, it requires a large workload, and often lead to deviation due to reader’s subjectivity. Therefore, based on automated segmentation of six horns of meniscus using fully and weakly supervised networks, we established two-layer cascaded classification models that can detect the meniscal lesions and further classify them into three types, and finally achieved excellent performance. This can improve the efficiency and accuracy of using quantitative MRI to study KOA in the future. |
| 4521 | Computer 91
|
Finite element simulation of breast hyperelasticity to develop a non-rigid registration model by deep learning |
| Fares OUADAHI1, Aurélie PETER1, Anaïs BERNARD1, and Julien ROUYER1 | ||
1Research and Innovation Department, Olea Medical, La Ciotat, France |
||
Keywords: Breast, Data Processing, Registration, Motion correction This work is part of an ongoing study to correct accidental breast motion during dynamic contrast-enhanced imaging. Here, we explored a finite element method to generate “moving” images based on a “fixed” one. The proposed methodology can help in the constitution of realistic dataset with large variety of motion to support the development of a dedicated non-rigid registration model by deep learning. |
| 4522 | Computer 92
|
Ensemble Multi-Path U-net Segmentation Algorithm for Breast Lesion based on Multi-Modality Image |
| Hang Yu1, Zichuan Xie2, Lizhi Xie3, Zhiheng Liu1, Lina Zhang4, Siyao Du4, Xiangjie Yin1, Chenyang Li1, Wenhong Jiang4, Yuru Guo1, and Zhongqi Kang4 | ||
1School of Aerospace Science and Technology, Xidian University, Xi'an, China, 2Guangzhou institute of technology, Xidian University, Guangzhou, China, 3GE Healthcare, Beijing, China, 4Department of Radiology, The First Hospital of China Medical University, Shenyang, China |
||
Keywords: Breast, Machine Learning/Artificial Intelligence, Deep Learning The multimodal MRI data is often ultilized for breast cancer analysis, and by now still difficult and inefficient to explored by segmentation algorithms. In this paper, we propose a MP-Unet based on U-net convolutional neural network, which can effectively ensemble multiple inputs of modal data and obtain accuracy segmentation results at the end. In MP-Unet, we reused some good quality modal data for training. The multiple MP-Unet models are further integrated based on Bagging algorithm to improve the segmentation accuracy of lesions. Experiments suggest that our proposed method has a huge performance improvement. |
| 4523 | Computer 93
|
Volumetric vocal tract segmentation using a deep transfer learning 3D U-Net model |
| Subin Erattakulangara1, Sarah Gerard1, Karthika Kelat1, Katie Burnham2, Rachel Balbi2, David Meyer2, and Sajan Goud Lingala1,3 | ||
1Roy J Carver Department of Biomedical Engineering, University of Iowa, iowa city, IA, United States, 2Janette Ogg Voice Research Center, Shenandoah University, Winchester, IA, United States, 3Department of Radiology, University of Iowa, iowa city, IA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Head & Neck/ENT The investigation into the 3D airway area is the prerequisite step for quantitatively studying the anatomical structures and function of the upper airway. Segmentation of upper airway can be considered as one of the stepping stones for this investigation. In this work, we propose a transfer learning-based 3D U-Net model with a ResNet encoder for vocal tract segmentation with small datasets training. We demonstrate its utility on sustained volumetric vocal tract MR scans from the recently released French speaker database. |
| 4732 | Computer 141
|
Deep learning-based image quality and spatial resolution improvement in Turbo Spin Echo Diffusion Weighted Imaging for prostate |
| Jihun Kwon1, Kohei Yuda2, Masami Yoneyama1, Yasutomo Katsumata3, and Marc Van Cauteren3 | ||
1Philips Japan, Tokyo, Japan, 2Tokyo Metropolitan Police Hospital, Nakano, Japan, 3Philips Healthcare, Best, Netherlands |
||
Keywords: Prostate, Diffusion/other diffusion imaging techniques Diffusion-weighted imaging (DWI) plays an important role in assessing the significance of prostate cancer. DWI with Turbo Spin Echo readout (TSE-DWI) is robust to image distortion but suffers from low signal to noise ratio. In this study, we investigated the use of prototype AI-based reconstruction technique (SmartSpeed Precise Image) to improve the image quality of TSE-DWI images. The image quality was compared between conventional Compressed-SENSE (C-SENSE), SmartSpeed AI, and SmartSpeed Precise Image. Volunteer data demonstrated a significant improvement of sharpness in both b=0 and 1000 s/mm2 images as well as ADC map, compared with C-SENSE and SmartSpeed AI reconstructions. |
| 4733 | Computer 142
|
Quantitative Analysis and Scan Time Impact of DL Recon Applied to Single-Shot Diffusion of the Prostate |
| Eugene Milshteyn1, Arnaud Guidon1, and Mukesh G. Harisinghani2 | ||
1GE Healthcare, Boston, MA, United States, 2Massachusetts General Hospital, Boston, MA, United States |
||
Keywords: Prostate, Diffusion/other diffusion imaging techniques Increasing the speed of multiparametric prostate MRI (mpMRI) is highly desirable. However, usual tradeoffs between signal-to-noise (SNR) and scan time must be considered and impact on quantitative metrics must be analyzed. One recently proposed approach applied a commercialized deep learning reconstruction (DL Recon) to prostate T2-weighted imaging, leveraging the capabilities of the DL algorithm to achieve a robust, high-quality T2-weighted acquisition in half the time. As such, this work focuses on evaluating the DL Recon on diffusion weighted imaging, which shows promise to cut acquisition time by ~70% and therefore benefit mpMRI. |
| 4734 | Computer 143
|
Deep learning-based high-resolution T2-weighted imaging elevated clarity of prostatic calcification and diagnostic efficacy |
| Zan Ke1, Liang Li1, Zhi Wen1, Weiyin Vivian Liu2, and Yunfei Zha1 | ||
1Radiology, Renmin Hospital of Wuhan University, Wuhan, China, 2GE Healthcare, MR Research China, Beijing, China |
||
Keywords: Prostate, Prostate Prostatic calcification is common in benign prostatic hyperplasia (BPH) and usually asymptomatic. Our study showed the prostate T2WIDL images have higher subjective rating scores, clearer lesion contrast and improved detection rate of prostatic calcification, higher SNR and CNR. In addition, T2WIDL more clearly and sharply displayed prostate capsule, lesion contrast, prostate calcification and anatomical details Therefore, AIR™ Recon DL based T2WI (T2WIDL) quality in prostate MRI offer higher overall image quality and elevated a younger radiologist’s diagnostic performance on prostate calcification. |
| 4735 | Computer 144
|
Conventional χ-separation compared to self calibrated method (BIOPHYSICSS-DL) with histological validation |
| Ilyes Benslimane1, Günther Grabner2, Simon Hametner3, Thomas Jochmann1,4, Robert Zivadinov1,5, and Ferdinand Schweser1 | ||
1Department of Neurology, Buffalo Neuroimaging Analysis Center, Buffalo, NY, United States, 2Department of Medical Engineering, Carinthia University of Applied Sciences, Klagenfurt, Austria, 3Department of Neuropathology and Neurochemistry, Medical University of Vienna, Vienna, Austria, 4Department of Computer Science and Automation, Technische Universität Ilmenau, Ilmenau, Germany, 5Department of Computer Science and Automation, Center for Biomedical Imaging, Clinical and Translational Science Institute at the University at Buffalo, Buffalo, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Brain The χ-separation method determines para- and diamagnetic susceptibility tissue compartments correlating to iron and myelin in the brain respectively. The method presupposes subject invariant relaxometry coefficients and compartments disregarding the changes in those parameters in disease or postmortem cases. We implement a biophysically informed autoencoder network developed for single subject use (BIOPHYSICSS-DL) to determine underlying biophysical model coefficients from individual datasets. We expand the current model with different combinations of relaxometry and susceptibility data to produce a self-calibrated χ separation method finding the network comparable to standard methods for iron and predicts myelin distribution more closely to ground truth histology. |
| 4736 | Computer 145
|
Motion artifact assessment for magnetic resonance imaging using a learned combination of deep learning model predictions. |
| Vanya Saksena1,2, Silvia Arroyo-Camejo2, Julian Hossbach1,2, Rainer Schneider2, and Andreas Maier1 | ||
1Pattern Recognition Lab, Friedrich-Alexander-University Erlangen-Nuremberg, Erlangen, Germany, 2Siemens Healthineers, Erlangen, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Artifacts, Image quality Magnetic resonance imaging (MRI) is a powerful imaging modality, but susceptible to various image quality problems. Today, technicians conduct image quality assurance (IQA) during scan-time as a manual, time-consuming, and subjective process. We propose a method towards adaptable automated IQA of MR images without the need of a large, annotated image database for training. Our method implements a machine learning-based module that uses multiple predictions from an ensemble of deep learning models trained with image quality metrics. The sensitivity of this method to detect image quality problems is adaptable to clinical requirements of the end user. |
| 4737 | Computer 146
|
Implicit Temporal-compensated Adversarial Network for 4D-MRI Enhancement |
| Yinghui Wang1, Tian Li1, Haonan Xiao1, and Jing Cai1 | ||
1Department of Health Technology and Informatics, The Hong Kong Polytechnic University, Hong Kong, Hong Kong |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, 4D-MRI\Enhancement\Temporal-compensation In this study, we proposed and evaluated a deep learning technique for refining four-dimensional magnetic resonance imaging (4D-MRI) in the post-processing stage. More specifically, we designed an implicit temporal-compensated adversarial network (ITAN) based on the intrinsic property of 4D-MRI to improve image quality with neighboring phases. It can overcome its inherent challenges of data deficiency, misalignment between training pairs, and complex texture details. The qualitative and quantitative results demonstrated that the proposed model can suppress the noise and artifacts in 4D-MR images, recover the missing details and perform better than a state-of-the-art method. |
| 4738 | Computer 147
|
Deep learning tools to aid the evaluation of isthmic spondylolysis |
| Vadim Malis1, Suraj Achar2, Dosik Hwang3, and Won C. Bae1,4 | ||
1Radiology, University of California, San Diego, La Jolla, CA, United States, 2Family Medicine, University of California, San Diego, La Jolla, CA, United States, 3Electrical and Electronics Engineering, Yonsei University, Seoul, Korea, Republic of, 4VA San Diego Healthcare System, San Diego, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Bone, UTE, Deep Learning, Image Regression, Saliency Map Isthmic spondylolysis results in fracture of pars interarticularis of the lumbar spine in young athletes. UTE MRI provide good bone contrast, although CT is still the gold standard. To take UTE MRI further, we developed supervised deep-learning tools to generate CT-like images and saliency maps of fracture probability from UTE MRI and CT, using ex vivo preparation of cadaveric spines. The results demonstrate feasibility of CT-like images to provide easier interpretability for bone fractures, due to improved image contrast and CNR, and the saliency maps to aid in quick detection of pars fracture by providing visual cues to the reader. |
| 4739 | Computer 148
|
A Fat-suppression, Image-subtraction Method Using Deep Learning for the Detection of Knee Abnormalities on MRI |
| Tsutomu Inaoka1, Akihiko Wada2, Tomoya Nakatsuka1, Masayuki Sugeta1, Akinori Yamamoto1, Hisanori Tomobe1, Ryousuke Sakai1, Hiroyuki Nakazawa1, Masaru Sonoda3, Rumiko Ishikawa1, Shusuke Kasuya1, and Hitoshi Terada1 | ||
1Radiology, Toho University Sakura Medical Center, Sakura, Japan, 2Radiology, Juntendo University, Tokyo, Japan, 3Radiology, Seirei Sakura Citizen Hospital, Sakura, Japan |
||
Keywords: Machine Learning/Artificial Intelligence, Joints This fat-suppression subtraction-image method using a DL model with 2D CNNs may be useful for the detection and classification of abnormalities on knee MRI. |
| 4740 | Computer 149
|
Accelerating DCE-MRI Analysis for Prostate Cancer Diagnosis with Deep Neural Networks |
| Kai Zhao1, Haoxin Zheng1, and Kyunghyun Sung1 | ||
1Department of Radiological Sciences, University of California, Los Angeles, Los Angeles, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Data Analysis, dynamic contrast enhanced A deep learning based DCE-MRI analysis method was proposed with a dedicated neural network architecture and data generation framework. The proposed method does not need DCE-MRI data acquisition or annotation for training. Compared to conventional non-linear least square (NLLS) fitting methods, the proposed method significantly reduced the average processing time from hours to few minutes while preserved the estimation quality. |
| 4741 | Computer 150
|
Distinction of True-Progression from Pseudo-progression in Glioblastomas using ML Model based on Quantitative mpMRI and Molecular Signatures |
| Virendra Kumar Yadav1, Suyash Mohan2, Sumeet Kumar Agarwal3,4, Laiz Laura de Godoy2, Sumei Wang2, MacLean P. Nasrallah5, Donald M. O’Rourke6, Stephen Bagley7, Harish Poptani8, Sanjeev Chawla2, and Anup Kumar Singh1,4,9 | ||
1Centre for Biomedical Engineering, Indian Institute of Technology, Delhi, India, 2Departments of Radiology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States, 3Department of Electrical Engineering, Indian Institute of Technology, Delhi, India, 4Yardi School of Artificial Intelligence, Indian Institute of Technology, Delhi, India, 5Clinical Pathology and Laboratory, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States, 6Neurosurgery, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States, 7Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, United States, 8Department of Molecular and Clinical Cancer Medicine, University of Liverpool, United Kingdom, Liverpool, United Kingdom, 9Department of Biomedical Engineering, All India Institute of Medical Sciences, New Delhi, India |
||
Keywords: Machine Learning/Artificial Intelligence, Brain Glioblastoma patients (n=93) exhibiting enhancing lesions within 6 months after completion of standard therapy underwent anatomical imaging, diffusion and perfusion MRI. The median values of parameters (MD, FA, CL, CP, CS and rCBV) were computed from the enhancing regions. O6-methylguanine-DNA-methyltransferase (MGMT) promoter methylation status was available from 75 patients. Subsequently, these patients were classified as TP (n=55) or PsP (n=20). The data were randomly split into training and testing sets. The best model for differentiating TP from PsP was obtained using quadratic SVM classifier with a training accuracy of 90.9%, cross-validation accuracy of 85.5% and testing accuracy of 85%. |
| 4742 | Computer 151
|
Anatomy and Image Contrast Metadata Verification using Self-supervised Pretraining |
| Ben A Duffy1 and Ryan Chamberlain1 | ||
1Subtle Medical Inc., Menlo Park, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Data Processing Automated metadata verification is essential for data quality checking. This study evaluates the extent to which self-supervised pretraining can improve performance on the anatomy and image contrast metadata verification tasks. On a small but diverse dataset, pretraining coupled with supervised finetuning, outperforms training from an ImageNet initialization, suggesting improved near out-of-distribution performance. On a larger brain-only dataset, training a linear classifier on the self-supervised pretrained network embeddings outperforms the corresponding ImageNet pretraining or random initialization on the image contrast prediction task. Cross-checking predictions against the DICOM metadata was an effective method for detecting artifacts and other quality issues. |
| 4743 | Computer 152
|
A neural network to estimate the hip center of rotation for a fully-automated range of motion analysis in femoroacetabular impingement. |
| Eros Montin1,2, Daniele Panozzo3, and Riccardo Lattanzi1,2,4 | ||
1Center for Advanced Imaging Innovation and Research (CAI2R) Department of Radiology, Radiology Department, New York University Grossman School of Medicine, New York, New York, USA, New York, NY, United States, 2Bernard and Irene Schwartz Center for Biomedical Imaging, Department of Radiology, New York University Grossman School of Medicine, New York, New York, USA, New York, NY, United States, 3New York University, New York, New York, USA, BROOKLYN, NY, United States, 4Vilcek Institute of Graduate Biomedical Sciences, New York University Grossman School of Medicine, New York, New York, USA, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Joints We evaluated three neural network architectures for the automatic identification of the center of the femur head on 3D water-only Dixon MRI. We trained using a mixture of real and augmented data. The mean error of the best-performing network was three-time lower compared to a manual annotation and on the order of 1 voxel. We combined the network to create the first fully automated pipeline to assess the hip range of motion from 3D MR. |
| 4744 | Computer 153
|
Can machine learning resolve model degeneracy in tissue microstructure estimation? |
| Michele Guerreri1,2, Sean Epstein1, Hojjat Azadbakht2, and Hui Zhang1 | ||
1Computer Science & Centre for Medical Image Computing, University College London, London, United Kingdom, 2AINOSTICS Ltd., Manchester, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence This work investigates the impact of model degeneracy on machine learning-based tissue microstructure estimation. While there have been several empirical reports suggesting machine learning can not resolve model degeneracy, the impact of model degeneracy is poorly understood. Here we show how model degeneracy can be categorised into three types with varying degrees of impact on machine learning-based microstructure estimation. Our finding is important for designing optimal training data distribution. |
| 4745 | Computer 154
|
A Deep Learning Nomogram Based on Gd-EOB-DTPA MRI for Predicting Early Recurrence in Hepatocellular Carcinoma after Hepatectomy |
| Meng Yan1, Xiao Zhang2, Bin Zhang1, Zhendong Qi3, Xiaoyun Liang2, Feng Huang2, Shuixing Zhang1, Xinming Li3, Shutong Wang4, and Xianyue Quan3 | ||
1Department of Radiology, The First Affiliated Hospital of Jinan University, Guangzhou, China, 2Neusoft Medical Systems Co., Ltd, Shanghai, China, 3Department of Radiology, Zhujiang Hospital of Southern Medical University, Guangzhou, China, 4Department of Liver Surgery, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China |
||
Keywords: Machine Learning/Artificial Intelligence, Cancer Prognostic risk assessment after hepatectomy for patients with hepatocellular carcinoma (HCC) remains difficult. Previous studies have shown that Gd-EOB-DTPA MRI is sensitive and accurate for HCC detection, but studies in predicting early recurrence after hepatectomy based on deep learning (DL) are still lacking. This study investigated the performance of a Gd-EOB-DTPA MRI-based DL approach, and then evaluated the DL nomogram incorporating deep features and significant clinical indicators. DL nomogram outperformed the clinical nomogram (validation AUC: 0.909 vs. 0.715). The proposed DL nomogram could provide a noninvasive and comprehensive tool for predicting early recurrence of HCC after curative resection. |
| 4746 | Computer 155
|
Improving Across-Dataset Schizophrenia Classification with Structural Brain MRI Using Multi-scale Transformer |
| Ye Tian1, Junhao Zhang2, Vish Mitnala Rao1, and Jia Guo3 | ||
1Biomedical Engineering, Columbia University, New York, NY, United States, 2BME, Columbia University, New York, NY, United States, 3Department of Psychiatry, Columbia University, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Brain, Deep Learning Schizophrenia is a neurological disorder that requires accurate and rapid detection for earlier intervention. Previous explorations in artificial intelligence showed overwhelming performance using deep learning in schizophrenia classification, though the generalization remained a challenge. We propose our 3D Multi-scale Transformer (MST) using T1W structural MRI data to detect schizophrenia. By synthesizing reconstructed images at different scales, the transformer-based architecture improves robustness to generalize in unseen data. The proposed method reaches the same-level performance of AUROC to the benchmark mark model in schizophrenia identification, and performs better in all leave-one-site-out generality tests. |
| 4747 | Computer 156
|
Using machine learning to evaluate values of six diffusion models to predict the efficacy of neoadjuvant chemotherapy for esophageal cancer |
| Long Cui1, Bingmei Bai2, Chenglong Wang1, Yang Song3, Shengyong Li1, Haijie Wang1, Jinrong Qu2, and Guang Yang1 | ||
1Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, Shanghai, China, 2Department of Radiology, Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital, Zhengzhou, China, 3MR Scientific Marketing, Siemens Healthcare, Shanghai, China |
||
Keywords: Cancer, Tumor We assessed the performance of diffusion models for assessing response to neoadjuvant chemotherapy (NACT) using machine learning. Firstly, features were extracted from the region of interest on different parametric maps of different diffusion models for esophageal squamous cell carcinoma (ESCC) patients and changes of the parameters (Δ parameter) before and after NACT (pre-NACT and post-NACT) were calculated. Then different Δ-NACT models and pre-NACT models were built for using features from different diffusion models. The results demonstrated that diffusion models may be used to predict the efficacy of NACT in ESCC patients. |
| 4748 | Computer 157
|
Multi-parametric MRI Radiomics Combined with Clinical Factors to Predict the Response to Neoadjuvant Chemotherapy in Nasopharyngeal Carcinoma |
| Zhuo Wang1, Zhiqiang Chen2, Xiaohua Chen1, Shaoru Zhang1, Yunshu Zhou1, Ruodi Zhang1, Shili Liu1, Yuhui Xiong3, and Bing chen4 | ||
1Department of Clinical Medicine of Ningxia Medical University, Yinchuan, China, 2Department of Radiology, the First Hospital Affiliated to Hainan Medical College, Haikou, China, 3GE Healthcare, Beijing, China, 4General Hospital of Ningxia Medical University, Yinchuan, China |
||
Keywords: Cancer, Head & Neck/ENT, multi-parametric MRI-based radiomics This study aims to evaluate the predictive performance of the nomogram integrating multi-parametric MRI-based radiomics and clinical characteristics in detecting therapeutic response to neoadjuvant chemotherapy (NAC) in nasopharyngeal carcinoma (NPC) patients. Least absolute shrinkage and selection operator (LASSO) regression was performed to select radiomics features. A nomogram was constructed using multivariable logistic regression. The receiver operating characteristic (ROC) curves, calibration, and decision curves were performed to assess the discriminative performance of the clinical, radiomics, and the combined models. The nomograms developed by integrating radiomics score (Rad-Score) with clinical factors outperformed the clinical model alone. |
| 4749 | Computer 158
|
Multi-parametric MRI-based Radiomics integrated with Clinical Features for Predicting the Ki-67 Labeling Index in Nasopharyngeal Carcinoma |
| Zhuo Wang1, Zhiqiang Chen2, Xiaohua Chen1, Shaoru Zhang1, Shili Liu1, Ruodi Zhang1, Yunshu Zhou1, Yuhui Xiong3, and AiJun Wang4 | ||
1Department of Clinical Medicine of Ningxia Medical University, Yinchuan, China, 2the First Hospital Affiliated to Hainan Medical College, Haikou, China, 3GE Healthcare, Beijing, China, 4General Hospital of Ningxia Medical University, Yinchuan, China |
||
Keywords: Cancer, Head & Neck/ENT, multi-parametric MRI-based radiomics To investigate the value of a nomogram based on multi-parametric MRI-based radiomics combined with clinical imaging features in predicting the Ki-67 LI in nasopharyngeal carcinoma. Least absolute shrinkage and selection operator (LASSO) regression was performed to select radiomics features. A nomogram was established using multivariable logistic regression. The receiver operating characteristic (ROC) curves, calibration, and decision curves were performed to evaluate the predictive performance of the different models. The nomograms constructed by integrating radiomics score (Rad-Score) with clinical imaging factors outperformed the clinical or the radiomics models alone. |
| 4750 | Computer 159
|
Multi-branch deep learning model integrated with multi-modal MRI for differential diagnosis of autism spectrum disorders |
| Xuan Yu1 and Meiyun Wang1,2 | ||
1Henan Provincial People’s Hospital & the People’s Hospital of Zhengzhou University, Zhengzhou, China, 2Laboratory of Brain Science and Brain-Like Intelligence Technology, Institute for Integrated Medical Science and Engineering, Henan Academy of Sciences, Zhengzhou, China |
||
Keywords: Brain Connectivity, Brain Connectivity This study proposes a multi-branch deep learning model fused with multi-modal MRI for the differential diagnosis of ASD. We solve the problem that traditional ASD classification algorithms based on static functional network connections ignore the time-varying characteristics of brain functional connections. Study the spatiotemporal characteristics of ASD brain imaging, and mine the information of functional connectivity between brain regions over time. |
| 4907 | Computer 141
|
Automatic localization of metal artifacts regions on MRI scout images |
| Deepa Anand1, Dattesh Shanbhag1, Chitresh Bhushan2, Kavitha Manickam2, and RAdhika Madhavan2 | ||
1GE Healthcare, Bangalore, India, 2GE Healthcare, Niskayuna, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Artifacts, Metal implants In MRI, presence of metal can cause blooming artifacts on MRI images. Automatically detecting the presence of metal and localizing it on the scout images itself can help streamline the MRI imaging workflow decisions downstream. In this work, we demonstrate a DL framework for hot spotting of metal/metal affected regions based on 2D three-plane MRI scout images of spine. The results indicate that the solution not only detects metal regions well on spine images (on which it was trained), but also is generalizable enough to work on other anatomies such as knee which was not part of the training data. |
| 4908 | Computer 142
|
Do deep learning-based qMRI parameter estimators improve clinical task performance? |
| Sean C. Epstein1,2, Timothy J. P. Bray3,4, Margaret Hall-Craggs3,4, and Hui Zhang1,2 | ||
1Computer Science, UCL, London, United Kingdom, 2Centre for Medical Image Computing, UCL, London, United Kingdom, 3Centre for Medical Imaging, UCL, London, United Kingdom, 4Imaging, UCLH, London, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Data Analysis We compare modern deep learning (DL)-based parameter-estimation methods to their traditional maximum-likelihood estimation (MLE) counterparts by evaluating each approach’s performance in two clinical classification tasks. This is motivated by recent work demonstrating the inherent bias-variance trade-off that differentiates different DL-based approaches. Results show how these trade-offs manifest in the ‘real world’ of tissue classification, and how they compare to the performance achievable with conventional iterative MLE. |
| 4909 | Computer 143
|
Quantitative Assessment of the Whole Spine in T2 MRI Using Deep Learning |
| Siavash Khallaghi1, Lucas Porto1, Sean London2, Yosef Chodakiewitz2, Rajpaul Attariwal1, and Sam Hashemi1 | ||
1Voxelwise Imaging Technology Inc., Vancouver, BC, Canada, 2Prenuvo, Vancouver, BC, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Spine, spondyloarthropathy, spondylosis We present a fully automatic system for the quantitative assessment of discs and vertebrae using convolutional neural networks. The proposed algorithm works in three stages: 1) segmentation/identification of spinal anatomy; 2) curvature analysis; and 3) detection of pathological conditions of intervertebral discs. We validate the proposed approach on a large dataset of 1,500 subjects with sagittal T2-weighted whole spine MRI, obtained as part of a whole body MRI protocol in a preventative health screening program. |
| 4910 | Computer 144
|
Classification of Plaque Composition in Peripheral Arterial Disease with Multi-Contrast Histology using 7 Tesla and a Variational AutoEncoder |
| Christof Karmonik1, Lily Buckner2, Kayla Wilhoit1, and Trisha Roy3 | ||
1Houston Methodist Research Institute, Houston, TX, United States, 2Houston Methodst Research Institute, Houston, TX, United States, 3Houston Methodist Hospital, Houston, TX, United States |
||
Keywords: Machine Learning/Artificial Intelligence, High-Field MRI, ultra-short TE, 7 Tesla To differentiate soft versus hard peripheral arterial disease (PAD) plaques, a dedicated ultra-high field magnetic resonance imaging (MRI) histology protocol using three different image contrast (ultra-short TE, T1-weighted and T2-weighted) was used to scan on a clinical 7 Tesla human MRI scanner six amputated legs immediately after surgery that were harboring PAD lesions. A 2D convolutional network (CNN) variational autoencoder (VAE) was created. For each lesion, average classification values along the long axis of the artery were determined from latent space and ranged from partially occluded to patent (three lesions), soft-plaque occluded (two) and hard-plaque occluded (one). |
| 4911 | Computer 145
|
Automatic Fetal Orientation Detection Algorithm in Fetal MRI |
| Joshua Eisenstat1, Matthias Wagner2, Logi Vidarsson3, Birgit Ertl-Wagner2,4, and Dafna Sussman1,5,6 | ||
1Department of Electrical, Computer and Biomedical Engineering, Faculty of Engineering and Architectural Sciences, Toronto Metropolitan University, Toronto, ON, Canada, 2Division of Neuroradiology, The Hospital for Sick Children, Toronto, ON, Canada, 3Department of Diagnostic Imaging, The Hospital for Sick Children, Toronto, ON, Canada, 4Department of Medical Imaging, Faculty of Medicine, University of Toronto, Toronto, ON, Canada, 5Institute for Biomedical Engineering, Science and Technology (iBEST), Toronto Metropolitan University and St. Michael’s Hospital, Toronto, ON, Canada, 6Department of Obstetrics and Gynecology, Faculty of Medicine, University of Toronto, Toronto, ON, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Convolutional Neural Networks, Fet-Net Fetal orientation determines the mode of delivery. It is also important for sequence planning in fetal MRI. This abstract proposes Fet-Net, a deep-learning algorithm, which uses a novel convolutional neural network (CNN) architecture, to automatically detect fetal orientation from a 2-dimensional (2D) magnetic resonance imaging (MRI) slice. 6,120 2D MRI slices displaying vertex, breech, oblique and transverse fetal orientations were used for training, validation and testing. Fet-Net achieved an average accuracy and F1 score of 97.68%, and a loss of 0.06828. Fet-Net was able to detect and classify fetal orientation, which may serve to accelerate fetal MRI acquisition. |
| 4912 | Computer 146
|
Deep Learning Based Real-time Quality Assessment of Pilot Tone Respiratory Signals |
| Huixin Tan1 and Yantu Huang1 | ||
1Siemens Shenzhen Magnetic Resonance Ltd., China, Shenzhen, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Pilot Tone (PT) respiratory signals are susceptible to interference like patient bulk motion or radio frequency interference. The signal curve is prone to bad when remains strong interference after suppression processing in the learning phase, resulting in low triggering accuracy and efficiency. With real-time quality assessment in the learning phase, a good-quality signal can be selected to learn better processing parameters. To ensure robustness and inference time, a tiny CNN is utilized to classify the signal into good and bad quality. Experimental results demonstrate that the proposed method has a strong ability to assess the quality of PT respiratory signals. |
| 4913 | Computer 147
|
Automated Motion Artifact Detection in Early Pediatric Diffusion MRI Using a Convolutional Neural Network |
| Jayse Merle Weaver1,2, Marissa DiPiero2,3, Patrik Goncalves Rodrigues2, Hassan Cordash2, and Douglas C Dean III1,2,4 | ||
1Medical Physics, University of Wisconsin-Madison, Madison, WI, United States, 2Waisman Center, University of Wisconsin-Madison, Madison, WI, United States, 3Neuroscience Training Program, University of Wisconsin-Madison, Madison, WI, United States, 4Pediatrics, University of Wisconsin-Madison, Madison, WI, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Artifacts, Quality Control A three-dimensional convolutional neural network was trained to detect motion artifacts on a volume level for two pediatric diffusion MRI datasets acquired between 1 month and 3 years of age. Accuracies of 95% and 98% were achieved between the two datasets. Additionally, the effects of motion-corrupted volumes on quantitative parameter estimation was examined. Data was processed without quality control and with quality control performed by the neural network. DTI and NODDI metrics were calculated and compared between methods. Significant differences were found for both individual and group results. |
| 4914 | Computer 148
|
Multimodal MRI radiomics for identifying true tumor recurrence and treatment-related effects in postoperative glioma patients |
| Jinfa Ren1, Dongming Han1, Xiaoyang Zhai1, Huijia Yin1, Ruifang Yan1, and Kaiyu Wang2 | ||
1Department of MR, The First Affiliated Hospital of Xinxiang Medical University, Weihui, China, 2GE Healthcare, MR Research China, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Radiomics Detecting true tumor recurrence and treatment-related effects in glioma after treatment is crucial for patient managements and challenging via conventional MRI for differentiation. Radiomics can be used to access the details in the images in an objective way. We constructed models based on multiple modalities by using radiomics features of the postoperative enhanced and edematous regions to find key features for identifying true tumor recurrence. Features from CE-T1WI and enhanced regions have excellent classification performance, and the model of multimodality with whole regions is the best, which may aid clinicians in developing individualized treatment strategies. |
| 4915 | Computer 149
|
An unsupervised deep learning-based method for in vivo high resolution Kidney MRI motion correction |
| Shahrzad Moinian1,2, Nyoman Kurniawan 1, Shekhar Chandra 3, Viktor Vegh1,2, and David Reutens1,2 | ||
1Centre for Advanced Imaging, The University of Queensland, St Lucia, Brisbane, Australia, 2Australian Research Council Training Centre for Innovation in Biomedical Imaging Technology, The University of Queensland, Brisbane, Australia, 3School of Information Technology and Electrical Engineering, The University of Queensland, Brisbane, Australia |
||
Keywords: Machine Learning/Artificial Intelligence, Motion Correction A primary challenge for in vivo kidney MRI is the presence of different types of involuntary physiological motion, affecting the diagnostic utility of acquired images due to severe motion artifacts. Existing prospective and retrospective motion correction methods remain ineffective when dealing with complex large amplitude nonrigid motion artifacts. We introduce an unsupervised deep learning-based method for in vivo kidney MRI motion correction. We demonstrate that our deep learning model achieved the average structural similarity index measure (SSIM) of 0.76±0.06 between the reconstructed motion-corrected and ground truth motion-free images, showing an improvement of about 0.33 compared to the corresponding motion-corrupted images. |
| 4916 | Computer 150
|
Trading Streamlines in Tractography using Autoencoders (TINTA) |
| Jon Haitz Legarreta1, Laurent Petit2, Pierre-Marc Jodoin1, and Maxime Descoteaux1 | ||
1Computer Science, Université de Sherbrooke, Sherbrooke, QC, Canada, 2Université Bordeaux, Bordeaux, France |
||
Keywords: Machine Learning/Artificial Intelligence, Tractography & Fibre Modelling Current tractography methods have a limited ability to accurately reconstruct the long-range brain white matter fiber pathways. Local orientation propagation methods provide tractograms with a non-negligible amount of implausible streamlines. In this work, we propose an artificial intelligence model to recover long-range white matter tracks that are potentially missed in conventional streamline propagation. Our method uses the generative ability of an autoencoder to propose new, plausible streamlines that are subsequently exchanged, according to a given similarity index, with the implausible streamlines in a given tractogram. This allows to potentially improve the reliability of the reconstructed fiber pathways. |
| 4917 | Computer 151
|
Region of Interest Prediction by Intra-stack Attention Neural Network |
| Ke Lei1, Ali B. Syed2, Xucheng Zhu3, John M. Pauly1, and Shreyas V. Vasanawala2 | ||
1Electrical Engineering, Stanford University, Stanford, CA, United States, 2Radiology, Stanford University, Stanford, CA, United States, 3GE Healthcare, Menlo Park, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, Region of Interest Manual prescription of the field of view (FOV) by MRI technologists is variable and prolongs the scanning process. Often, the FOV is too large or crops critical anatomy. We propose a deep-learning framework, trained by radiologists’ supervision, for predicting region of interest (ROI) and automating FOV prescription. The proposed ROI prediction model achieves an average IoU of 0.867, significantly better (P<0.05) than two baseline models and not significantly different from a radiologist (P>0.12). The FOV prescribed by the proposed framework achieves an acceptance rate of 92% from an experienced radiologist. |
| 4918 | Computer 152
|
Free water imaging parameter estimation by combination of synthetic q-space learning and conventional fitting: a hybrid approach |
| Keigo Yamazaki1,2, Yoshitaka Masutani3, Wataru Uchida1, Koji Kamagata1, Koh Sasaki4,5, and Shigeki Aoki1 | ||
1Department of Radiorogy, Juntendo University graduate School of Medicine, Tokyo, Japan, 2Department of Radiological Science, Tokyo Metropolitan University, Tokyo, Japan, 3Tohoku University graduate School of Medicine, Miyagi, Japan, 4Graduate School of Infomation Sciences, Hiroshima City University, Hiroshima, Japan, 5Hiroshima Heiwa Clinic, Hiroshima, Japan |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Free water imaging (FWl), among the diffusion MRI (dMRI) family, is an extended version of the single diffusion tensor model by adding the isotropic diffusion compartment. Generally, FWI parameters have been estimated by fitting of signal model to measured DWI signals. Recently, machine learning techniques have shown promising results also in dMRI parameter inference.In this study, we aimed at development of a hybrid approach for FWI parameter estimation based on synthetic q-space learning (synQSL) and conventional fitting.Our approach was validated by comparison with the conventional fitting method based on quantitative and visual evaluation and computation time. |
| 4919 | Computer 153
|
Transformer-based image quality improvement of radial undersampled lung MRI data from post-COVID-19 patients |
| Maximilian Zubke1,2, Robin A Müller1,2, Marius Wernz1,2, Filip Klimeš1,2, Frank Wacker1,2, and Jens Vogel-Claussen1,2 | ||
1Institute of Diagnostic and Interventional Radiology, Hannover Medical School, Hannover, Germany, 2Biomedical Research in Endstage and Obstructive Lung Disease Hannover (BREATH), German Center for Lung Research (DZL), Hannover, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Currently, MR-based ventilation imaging relying on radial 3D-stack-of-stars spoiled gradient echo sequence requires a fairly long acquisition time of 8 minutes, which may impact clinical translation. Therefore, a shorter acquisition time is desired. In this study, a novel deep learning approach called transformer was evaluated for image restauration of radial undersampled lung images from 16 post-COVID-19 patients. For each patient, images resulting from 4- and 8 minutes acquisitions were provided. A transformer was trained to translate the 4-minute-version to the corresponding 8-minute-version and led to a significant image quality improvement, demonstrated by three complementary image similarity metrics. |
| 4920 | Computer 154
|
Deep learning based estimation of patient anthropometric data for intelligent scan planning in MR |
| Subham Kumar1 and Harikrishna Rai1 | ||
1GE Healthcare, Bengaluru, India |
||
Keywords: Machine Learning/Artificial Intelligence, Data Analysis Estimating patient height and weight is used to determine the optimum SAR (Signal Modelling) for the patient. We used a deep learning approach to first generate the point cloud of the person (Data Processing) followed by prediction of the height and weight (Data Analysis). This tackles the problems of heavy occlusion which occurs in the MR imaging scenario in the form of coils/blankets and different positions in which the patient will be placed. We achieved a MAE score of 4.9 cm on the height and 6.2 kg on the weight. This is a promising solution to an important problem. |
| 4921 | Computer 155
|
Image Quality and Quantitative Analysis of abbreviated IVIM Brain MRI With Deep Learning–Based Reconstruction |
| Qiongge Li1, Yayan Yin1, and Jie Lu1 | ||
1Xuanwu Hospital Capital Medical University, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Diffusion/other diffusion imaging techniques Deep learning-based reconstruction may improve the image signal to noise ratio without impacting the image contrasts. Intravoxel incoherent motion (IVIM) often require multiple b values with multiple averages that give rise to prolonged scan time. In this work, deep learning reconstruction is used to reduce the overall IVIM scan time. Based on qualitative and quantitative analysis, deep learning reconstruction may significantly improve the results of IVIM and make an abbreviated IVIM feasible. |
| 4922 | Computer 156
|
Predicting Double Expression Status in Primary Central Nervous System Lymphoma Using Multiparametric MRI Based Machine Learning |
| Li Guo Liu1, Yue Xin Zhang1, Nan Zhang1, Feng Hua Xiao1, Jing xin Chen1, and Lin Ma1 | ||
1Department of Radiology, The First Medical Center, Chinese PLA General Hospital, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Multimodal In this study, we proposed a promising method to distinguish the double expression lymphoma (DEL) from the non-double expression lymphoma (non-DEL) in primary central nervous system lymphoma (PCNSL) by using multiparametric MRI-based machine learning. The results showed that clinical characteristics and MR imaging features had no significant differences in distinguishing DEL from non-DEL . However, radiomics features could differentiate the two status and the best model in this study was SVMlinear with the combined four sequence group (AUCmean = 0.89±0.04). So multiparametric MRI based machine learning is promising in predicting DEL status in PCNSL. |
| 4923 | Computer 157
|
Deep learning-based pipeline to improve sharpness in knee imaging at both 1.5T and 3T: a clinical evaluation |
| Valentin H Prevost1, Shelton Caruthers1, Khadra Fleury2, Wissam AlGhuraibawi3, and Kensuke Shinoda1 | ||
1Canon Medical Systems Corporation, Otawara, Japan, 2Canon Medical Systems France, Paris, France, 3Canon Medical Systems USA, Tustin, CA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence In MRI, edge sharpness is one of the main criteria to allow structure’s delineation and relevant clinical diagnosis. One way to improve the sharpness is to artificially increase the reconstructed matrix size with methods such as zero-padding interpolation (ZIP). In a previous study, we created a deep learning reconstruction (DLR) pipeline, combining ZIP with two CNN’s: the first one trained to reduce image noise, and the second one to reduce ringing artifacts. The goal of this work was to evaluate the clinical impact of this DLR pipeline on pathological knee images performed at 1.5T and 3T, compared to standard reconstructions. |
| 4924 | Computer 158
|
High Resolution Image Quality Improvement of T2 FLAIR Coronal Hippocampal Imaging by Deep Learning Reconstruction |
| Yang Jing1, Li Qiong Ge1, Wu Tao2, Qi Zhi Gang1, Zhao Cheng1, and Lu Jie1 | ||
1Department of Radiology and Nuclear Medicine, xuanwu hospital, Capital Medical University, Beijing, China, 2General Electric Medical System Trade Development (Shanghai) Co., Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Brain, Hippocampus, Deep learning reconstruction, Epilepsy Deep learning reconstruction (DL Recon) method can improve the image signal-to-noise ratio and contrast-to-noise ratio without prolonging the scanning time and affecting the signal intensity difference of bilateral hippocampus. It can improve the image quality of hippocampus toe and surrounding small lesions dramatically as in Fig.1. |
| 4925 | Computer 159
|
Association between myosteatosis and impaired glucose metabolism: A deep learning whole-body MRI population phenotyping approach |
| Matthias Jung1, Marco Reisert2, Susanne Rospleszcz3,4, Annette Peters3,4, Johanna Nattenmüller1, Christopher L. Schlett1, Fabian Bamberg1, and Jakob Weiss1 | ||
1Department of Diagnostic and Interventional Radiology, University Medical Center Freiburg, Freiburg, Germany, 2Medical Physics, Department of Diagnostic and Interventional Radiology, University Medical Center Freiburg, Freiburg, Germany, 3Institute of Epidemiology, Helmholtz Zentrum München, Neuherberg, Germany, 4Institute for Medical Information Processing, Biometry and Epidemiology, Ludwig-Maximilians-University München, München, Germany |
||
Keywords: Screening, Diabetes Diabetes remains a major challenge for healthcare systems, making screening and early detection desirable. We used deep learning to quantify myosteatosis as 1) skeletal muscle fat fraction (SMFF) and 2) intramuscular adipose tissue (IMAT) normalized for SM mass and assessed their association with impaired glucose metabolism. SMFF had a higher discriminatory capacity for impaired glucose metabolism than IMAT. In multivariable logistic regression adjusted for baseline demographics and cardiometabolic risk factors, only SMFF remained an independent predictor of impaired glucose metabolism. Deep learning-based MR phenotyping enables opportunistic screening of myosteatosis and may identify individuals at high risk for impaired glucose metabolism. |
| 4926 | Computer 160
|
Convolutional Neural Network based Stack-of-Star Imaging with Noise and Artifacts Removal |
| Xinzeng Wang1, Yedaun Lee2,3, Joonsung Lee4, Sagar Mandava5, Ty A. Cashen6, Xucheng Zhu7, and Arnaud Guidon8 | ||
1GE Healthcare, Houston, TX, United States, 2Department of Radiology, Haeundae Paik Hospital, Busan, Korea, Republic of, 3Inje University College of Medicine, Busan, Korea, Republic of, 4GE Healthcare, Seoul, Korea, Republic of, 5GE Healthcare, Atlanta, GA, United States, 6GE Healthcare, Madison, WI, United States, 7GE Healthcare, Menlo Park, CA, United States, 8GE Healthcare, Boston, MA, United States |
||
Keywords: Cancer, Image Reconstruction, Liver, Pancreas Stack of Star acquisition is one of the most frequently used non-Cartesian k-space sampling methods due to its fast speed and robustness to motion. To further reduce the scan time or increase the temporal resolution, Stack of Star is often down-sampled with fewer spokes and advanced sampling patterns, such as golden angle acquisition. However, this makes Stack of Star prone to noise and streak artifacts, limiting the in-plane resolution and degrading diagnostic quality. In this work, we evaluated a deep-learning based stack-of-star method for free-breathing abdominal imaging and it shows improved diagnostic quality. |
| 5026 | Computer 81
|
Deep learning-based APT imaging using synthetically generated training data |
| Malvika Viswanathan1, Leqi Yin2, Yashwant Kurmi1, and Zhongliang Zu3 | ||
1Vanderbilt University Medical Center, Vanderbilt University Institute of Imaging Sciences, Nashville, TN, United States, 2Vanderbilt University, Nashville, TN, United States, 3Department of Radiology and Radiological Sciences, Vanderbilt University Institute of Imaging Sciences, Nashville, TN, United States |
||
Keywords: Machine Learning/Artificial Intelligence, CEST & MT Machine learning is increasingly applied to address challenges in specifically quantifying APT effect. The models are usually trained on measured data, which, however are usually lack of ground truth and sufficient training data. Synthetically generated data from both measurements and simulations can create training data which mimic tissues better than full simulations, cover all possible variations in sample parameters, and provide the ground truth. We evaluated the feasibility to use synthetic data to train models for predicting APT effect. Results show that the machine learning predicted APT is more close to the ground truth than the conventional multiple-pool Lorentzian fit. |
| 5027 | Computer 82
|
Conditional VAE for Single-Voxel MRS Data Generation |
| Dennis van de Sande1, Sina Amirrajab1, Mitko Veta1, and Marcel Breeuwer1,2 | ||
1Biomedical Engineering - Medical Image Analysis Group, Eindhoven University of Technology, Eindhoven, Netherlands, 2MR R&D - Clinical Science, Philips Healthcare, Best, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Spectroscopy, deep learning, generative modelling We propose a conditional VAE to synthesize single-voxel MRS data. This deep learning method can be used to enrich in-vivo datasets for other machine learning applications, without using any physics-based models. Our work is a proof-of-concept study which demonstrates the potential of a cVAE for MRS data generation by using a synthetic dataset of 8,000 spectra for training. We evaluate our model by performing a linear interpolation of the latent space, which shows that spectral properties are captured in the latent space, meaning that our model can learn spectral features from the data and can generate new samples. |
| 5028 | Computer 83
|
Estimating uncertainty in diffusion MRI models using generative deep learning |
| Frank Zijlstra1,2 and Peter T While1,2 | ||
1Department of Radiology and Nuclear Medicine, St. Olav's University Hospital, Trondheim, Norway, 2Department of Circulation and Medical Imaging, NTNU - Norwegian University of Science and Technology, Trondheim, Norway |
||
Keywords: Machine Learning/Artificial Intelligence, Diffusion/other diffusion imaging techniques, Model fitting, uncertainty, IVIM Uncertainty is an important aspect of fitting quantitative models to diffusion MRI data, which is often overlooked. This study presents a method for estimating uncertainty intrinsic to a model using a generative deep learning approach (Denoising Diffusion Probabilistic Models (DDPM)). We numerically validate that the approach provides accurate uncertainty estimates, and demonstrate its use in providing signal-specific uncertainty estimates. Furthermore, we show that DDPM can be used as a fitting method that estimates uncertainty, and show both ADC and IVIM fitting on an in vivo brain scan. This shows promise for DDPM, as both an investigate tool and fitting method. |
| 5029 | Computer 84
|
Synthesized 7T MRI from 3T MRI using generative adversarial network: validation in clinical brain imaging |
| Caohui Duan1, Xiangbing Bian1, Kun Cheng1, Jinhao Lyu1, Yongqin Xiong1, Jianxun Qu2, Xin Zhou3, and Xin Lou1 | ||
1Department of Radiology, Chinese PLA General Hospital, Beijing, China, 2MR Collaboration, Siemens Healthineers Ltd., Beijing, China, 3Key Laboratory of Magnetic Resonance in Biological Systems, State Key Laboratory of Magnetic Resonance and Atomic and Molecular Physics, National Center for Magnetic Resonance in Wuhan, Wuhan Institute of Physics and Mathematics, Innovation Academy for Precision Measurement Science and Technology, Chinese Academy of Sciences‒Wuhan National Laboratory for Optoelectronics, Wuhan, China |
||
Keywords: Machine Learning/Artificial Intelligence, Brain Ultra-high field 7T MRI provides exceptional tissue contrast and anatomical details but is often cost-prohibitive and not widely accessible in clinics. A generative adversarial network (SynGAN) was developed to generate synthetic 7T images from the widely used 3T images. The synthetic 7T images achieved improved tissue contrast and anatomical details compared to the 3T images. Meanwhile, the synthetic 7T images showed comparable diagnostic performance to the authentic 7T images for visualizing a wide range of pathology, including cerebral infarction, demyelination, and brain tumor. |
| 5030 | Computer 85
|
Synthetising myelin water fraction from T1-weighted and T2-weighted data: an image-to-image translation approach |
| Matteo Mancini1, Carolyn McNabb1, Derek Jones1, and Mara Cercignani1 | ||
1Cardiff University Brain Research Imaging Centre, Cardiff University, Cardiff, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Brain Myelin biomarkers are a fundamental tool for both neuroscience research and clinical applications. Despite several quantitative MRI methods available for their estimation, in several cases qualitative approaches are the only viable solution. To get the best of both the quantitative and qualitative worlds, here we propose an image-to-image translation method to learn the mapping between common routine scans and a quantitative myelin metric. To achieve this goal, we trained a generative adversarial network on a relatively large dataset of healthy subjects. Both the qualitative and quantitative results show good agreement between the predicted and the ground-truth maps. |
| 5031 | Computer 86
|
Resilience of synthetic CT DL network to varying ZTE-MRI input SNR |
| Sandeep Kaushik1,2, Cristina Cozzini1, Mikael Bylund3, Steven Petit4, Bjoern Menze2, and Florian Wiesinger1 | ||
1GE Healthcare, Munich, Germany, 2Department of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland, 3Umeå University, Umeå, Sweden, 4Erasmus MC Cancer Institute, Rotterdam, Netherlands |
||
Keywords: Machine Learning/Artificial Intelligence, Radiotherapy, MR-only RT, Synthetic CT, Multi-task CNN, PET/MR Many recent works have proposed methods to convert MRI into synthetic CT (sCT). While they have demonstrated a certain level of accuracy, not many have studied the robustness of those methods. In this work, we study the robustness of a multi-task deep learning (DL) model that computes sCT images from fast ZTE MR images under different levels of image noise. We evaluate its impact on radiation therapy planning. The proposed method demonstrates resilience against input noise variations. It makes way for a clinically acceptable dose calculation with a fast input image acquisition. |
| 5032 | Computer 87
|
Deep learning-based synthesis of TSPO PET from T1-weighted MRI images only |
| Matteo Ferrante1, Marianna Inglese2, Ludovica Brusaferri3, Marco L Loggia3, and Nicola Toschi4,5 | ||
1Biomedicine and prevention, University of Rome Tor Vergata, Roma, Italy, 2Biomedicine and prevention, University of Rome Tor Vergata, Rome, Italy, 3Martinos Center For Biomedical Imaging, MGH and Harvard Medical School (USA), Boston, MA, United States, 4BioMedicine and prevention, University of Rome Tor Vergata, Rome, Italy, 5Department of Radiology,, Athinoula A. Martinos Center for Biomedical Imaging and Harvard Medical school, Boston, MA, USA, Boston, MA, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Neuroinflammation, PET, image synthesis Chronic pain-related biomarkers can be found using a specific binding radiotracer called [11C]PBR28 able to target the translocator protein (TSPO), whose expression is increased in activated glia and can be considered as a biomarker for neuroinflammation. One of the main drawbacks of PET imaging is radiation exposure, for which we attempted to develop a deep learning model able to synthesize PET images of the brain from T1w MRI only. Our model produces synthetic TSPO-PET images from T1W MRI which are statistically indistinguishable from the original PET images both on a voxel-wise and on a ROI-wise level. |
| 5033 | Computer 88
|
Improving Prostate Cancer Detection Using Bi-parametric MRI with Conditional Generative Adversarial Networks |
| Alexandros Patsanis1, Mohammed R. S. Sunoqrot1,2, Elise Sandsmark2, Sverre Langørgen2, Helena Bertilsson3,4, Kirsten Margrete Selnæs1,2, Hao Wang5, Tone Frost Bathen1,2, and Mattijs Elschot1,2 | ||
1Department of Circulation and Medical Imaging, Norwegian University of Science and Technology - NTNU, Trondheim, Norway, 2Department of Radiology and Nuclear Medicine, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway, 3Department of Clinical and Molecular Medicine, Norwegian University of Science and Technology - NTNU, Trondheim, Norway, 4Department of Urology, St. Olavs Hospital, Trondheim University Hospital, Trondheim, Norway, 5Department of Computer Science, Norwegian University of Science and Technology - NTNU, Gjøvik, Norway |
||
Keywords: Machine Learning/Artificial Intelligence, Data Analysis, Deep Learning This study investigated automated detection and localization of prostate cancer on biparametric MRI (bpMRI). Conditional Generative Adversarial Networks (GANs) were used for image-to-image translation. We used an in-house collected dataset of 811 patients with T2- and diffusion-weighted MR images for training, validation, and testing of two different bpMRI models in comparison to three single modality models (T2-weighted, ADC, high b-value diffusion). The bpMRI models outperformed T2-weighted and high b-value models, but not ADC. GANs show promise for detecting and localizing prostate cancer on MRI, but further research is needed to improve stability, performance and generalizability of the bpMRI models. |
| 5034 | Computer 89
|
AI-based mapping from MRI to MR thermometry for MR-guided laser interstitial thermal therapy using a conditional generative adversarial network |
| Saba Sadatamin1,2, Steven Robbins3, Richard Tyc3, Adam C. Waspe2,4, Lueder A. Kahrs1,5,6,7, and James M. Drake1,2,8 | ||
1Institute of Biomedical Engineering, University of Toronto, Toronto, ON, Canada, 2Posluns Centre for Image Guided Innovation & Therapeutic Intervention, Hospital of Sick Children, Toronto, ON, Canada, 3Monteris Medical, Winnipeg, MB, Canada, 4Department of Medical Imaging, University of Toronto, Toronto, ON, Canada, 5Medical Computer Vision and Robotics Lab, University of Toronto, Toronto, ON, Canada, 6Department of Mathematical & Computational Sciences, University of Toronto Mississauga, Toronto, ON, Canada, 7Department of Computer Science, University of Toronto, Toronto, ON, Canada, 8Department of Surgery, University of Toronto, Toronto, ON, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, MR-Guided Interventions, Laser Interstitial Thermal Therapy MR-guided laser interstitial thermal therapy (MRgLITT) is a minimally-invasive treatment for brain tumors where a surgeon inserts a laser fiber along a fixed trajectory. Repositioning the laser is invasive and predicting thermal spread close to heat sinks is difficult. To address this problem, MR thermometry prediction using artificial intelligence (AI) modeling will be developed to aid the surgeon to determine whether a selected laser position is ideal before the treatment starts. AI algorithms will be trained to model the nonlinear mapping from anatomical MRI planning images to MR thermometry. A surgeon will choose a better fiber trajectory by AI model. |
| 5035 | Computer 90
|
Prediction of NF2 Loss in Meningiomas Using T1-Weighted Contrast Enhanced MRI Generated by Deep Convolutional Generative Adversarial Networks |
| Sukru Samet Dindar1, Buse Buz-Yalug2, Kubra Tan3, Ayca Ersen Danyeli4,5, Ozge Can5,6, Necmettin Pamir5,7, Alp Dincer5,8, Koray Ozduman5,7, Yasemin P. Kahya1, and Esin Ozturk-Isik2,5 | ||
1Electrical and Electronics Engineering, Bogazici University, Istanbul, Turkey, 2Institute of Biomedical Engineering, Bogazici University, Istanbul, Turkey, 3Health Institutes of Turkey, Istanbul, Turkey, 4Department of Medical Pathology, Acibadem University, Istanbul, Turkey, 5Brain Tumor Research Group, Acibadem University, Istanbul, Turkey, 6Department of Medical Engineering, Acibadem University, Istanbul, Turkey, 7Department of Neurosurgery, Acibadem University, Istanbul, Turkey, 8Department of Radiology, Acibadem University, Istanbul, Turkey |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Neurofibromatosis type 2 (NF2) gene mutations have been linked to tumorigenesis in meningiomas. This study aims to improve the prediction of NF2 loss in meningiomas using T1-weighted contrast-enhanced MRI augmented by a deep convolutional generative adversarial network (DCGAN). Synthetically generated MRI increased the training accuracy from 78.9% to 93% and test accuracy from 69.4% to 79.5% in this study. |
| 5036 | Computer 91
|
Synthetic CT generation from different MR contrast inputs and evaluation of its quantitative accuracy |
| Sandeep Kaushik1,2, Cristina Cozzini1, Jonathan J Wyatt3, Hazel McCallum3, Ross Maxwell4, Bjoern Menze2, and Florian Wiesinger1 | ||
1GE Healthcare, Munich, Germany, 2Department of Quantitative Biomedicine, University of Zurich, Zurich, Switzerland, 3Newcastle University and Northern Centre for Cancer Care, Newcastle upon Tyne, United Kingdom, 4Newcastle University, Newcastle upon Tyne, United Kingdom |
||
Keywords: Machine Learning/Artificial Intelligence, Radiotherapy, Synthetic CT, radiation therapy planning, Multi-task network, deep learning MRI to synthetic CT image conversion is a problem of interest for clinical applications such as MR-radiation therapy planning, PET/MR attenuation correction, MR bone imaging. Many methods proposed for this purpose use different MR inputs. In this work, we compare the sCT generated from different 3D MR inputs, including Zero TE (ZTE), fast spin echo (CUBE), and fast spoiled gradient echo with Dixon-type fat-water separation (LAVA-Flex), using a multi-task deep learning (DL) model. We analyze the qualitative and quantitative accuracy of the generated sCT image from each input and highlight the aspects relevant for different clinical applications. |
| 5037 | Computer 92
|
Synthesising MRIs from CTs to Improve Stroke Treatment Using Deep Learning |
| Grace Wen1, Jake McNaughton1, Ben Chong1, Vickie Shim1, Justin Fernandez1, Samantha Holdsworth2,3, and Alan Wang1,2,3 | ||
1Auckland Bioengineering Institute, University of Auckland, Auckland, New Zealand, 2Faculty of Medical and Health Sciences, University of Auckland, Auckland, New Zealand, 3Mātai Medical Research Institute, Tairāwhiti-Gisborne, New Zealand |
||
Keywords: Machine Learning/Artificial Intelligence, Stroke, Image Synthesis MRI holds an important role in diagnosing brain conditions, however, many patients do not receive an MRI before their diagnosis and onset of treatment. We propose to use deep learning to generate an MRI from a patient's CT and have implemented multiple models to compare their results. Using CT/MRI pairs from 181 stroke patients, we use mutiple deep learning models to generate MRI from the CT images. The model produces high quality images and accurately translates lesions onto the target image. |
| 5038 | Computer 93
|
Synthetic data driven learning for full-automatic hemoperfusion parameter estimation |
| Lu Wang1, Pujie Zhang1, Zhen Xing2, Congbo Cai1, Zhong Chen1, Dairong Cao2, Zhigang Wu3, and Shuhui Cai1 | ||
1Department of Electronic Science, Xiamen University, Xiamen, China, 2Department of Radiology, First Affiliated Hospital of Fujian Medical University, Fuzhou, China, 3MSC Clinical & Technical Solutions, Philips Healthcare, Shenzhen, China |
||
Keywords: Machine Learning/Artificial Intelligence, Brain, Hemoperfusion parameter estimation Hemoperfusion magnetic resonance (MR) imaging derived parameters characterize both endothelial hyperplasia and neovascularization that are associated with tumor aggressiveness and growth. However, the hemoperfusion parameter estimation is still limited by low reliability, high bias, long processing time, and operator experience dependency up to now. In this study, a synthetic data driven learning method for hemoperfusion parameter estimation is proposed. Image analysis shows that the proposed method improves the reliability and precision of hemodynamic parameter estimation in a full-automatic and high-efficient way. |
| 5039 | Computer 94
|
Bony structure enhanced synthetic CT generation using Dixon sequences for pelvis MR-only radiotherapy |
| Xiao Liang1, Ti Bai1, Andrew Godley1, Chenyang Shen1, Junjie Wu1, Boyu Meng1, Mu-Han Lin1, Paul Medin1, Yulong Yang1, Steve Jiang1, and Jie Deng1 | ||
1Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence Synthetic CT (sCT) image generated from MRI by unsupervised deep learning models tends to have large errors around bone area. To generate better sCT image quality in bone area, we propose to add bony structure constrains in the loss function of the unsupervised CycleGAN model, and modify the single-channel CycleGAN to a multi-channel CycleGAN that takes Dixon constructed MR images as inputs. The proposed model has lowest mean absolute error compared with single-channel CycleGAN with different MRI images as input. We found that it can generate more accurate Hounsfield Unit and anatomy of bone in sCT. |
| 5040 | Computer 95
|
Amyloid-Beta Axial Plane PET Synthesis from Structural MRI: An Image Translation Approach for Screening Alzheimer’s Disease |
| Fernando Vega1,2,3, Abdoljalil Addeh1,2,3, Ahmed Elmenshawi2, and M. Ethan MacDonald1,2,3,4 | ||
1Biomedical Engineering, Schulich School of Engineering, University of Calgary, Calgary, AB, Canada, 2Electrical & Software Engineering, Schulich School of Engineering, University of Calgary, Calgary, AB, Canada, 3Hotchkiss Brain Institute, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada, 4Department of Radiology, Cumming School of Medicine, University of Calgary, Calgary, AB, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Alzheimer's Disease, MRI, PET, Image Translation In this work, an image translation model is implemented to produce synthetic amyloid-beta PET images from structural MRI that are quantitatively accurate. Image pairs of amyloid-beta PET and structural MRI were used to train the model. We found that the synthetic PET images could be produced with a high degree of similarity to truth in terms of shape, contrast and overall high SSIM and PSNR. This work demonstrates that performing structural to quantitative image translation is feasible to enable the access amyloid-beta information from only MRI. |
| 5041 | Computer 96
|
3D Lesion Generation Model considering Anatomic Localization to Improve Object Detection in Limited Lacune Data |
| Daniel Kim1, Jae-Hun Lee1, Mohammed A. Al-masni2, Jun-ho Kim1, Yoonseok Choi1, Eun-Gyu Ha1, SunYoung Jung3, Young Noh4, and Dong-Hyun Kim1 | ||
1Department of Electrical and Electronic Engineering, Yonsei Univ., Seoul, Korea, Republic of, 2Department of Artificial Intelligence, Sejong Univ., Seoul, Korea, Republic of, 3Department of Biomedical Engineering, Yonsei Univ., Wonju, Korea, Republic of, 4Department of Neurology, Gachon Univ., Incheon, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Data Processing, Augmentation Recently, research on the detection of cerebral small vessel disease (CSVD) has been mainly implemented in two-stages (1st: candidate detection, 2nd: false-positive reduction). Previous studies presented the difficulty of collecting labeled data as a limitation. Here, we synthesized the lesion through 3D-DCGAN and insert it at different locations on the MR image considering anatomical localization and alpha blending to augment labeled data. Through this, the detecting architecture was simplified to a single-stage, and the precision and recall values were improved by an average of 0.2. |
| 5042 | Computer 97
|
Use of an Automated Approach for Generating vADC for a Large Patient Population Studied with 129Xe MRI |
| Ramtin Babaeipour1, Maria Mihele2, Keeirah Raguram2, Matthew Fox2,3, and Alexei Ouriadov1,2,3 | ||
1School of Biomedical Engineering, The University of Western Ontario, London, ON, Canada, 2Department of Physics and Astronomy, The University of Western Ontario, London, ON, Canada, 3Lawson Health Research Institute, London, ON, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Segmentation, Deep learning, Transfer learning Hyperpolarized 129Xe lung MRI is an efficient technique used to investigate and assess pulmonary diseases. However, the longitudinal observation of the emphysema progression using hyperpolarized gas MRI-based ADC can be problematic, as the disease-progression can lead to increasing unventilated-lung areas, which likely excludes the largest ADC estimates. One solution to this problem is to combine static-ventilation and ADC measurements following the idea of 3He MRI ventilatory ADC (vADC). We have demonstrated this method adapted for 129Xe MRI to help overcome the above-mentioned shortcomings and provide an accurate assessment of the emphysema progression. |
| 5179 | Computer 81
|
Reproducible DL-based approach for liver PDFF quantification |
| Juan Pablo Meneses1,2,3, Cristobal Arrieta1,2, Pablo Irarrazaval1,3,4, Cristian Tejos1,3, Marcelo Andía1,2,5, Carlos Sing Long1,4,6, and Sergio Uribe1,2,5 | ||
1Biomedical Imaging Center, Pontificia Universidad Católica de Chile, Santiago, Chile, 2i-Health Millennium Institute for Intelligent Healthcare Engineering, Santiago, Chile, 3Department of Electrical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile, 4Institute for Biological and Medical Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile, 5Radiology Department, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile, 6Institute for Mathematical & Computational Engineering, Pontificia Universidad Católica de Chile, Santiago, Chile |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Imaging, Convolutional Neural Network Liver PDFF is a biomarker correlated with hepatic pathologies. Recently, several Deep Learning (DL) methods have been proposed to accelerate the necessary post-processing to estimate PDFF. However, none of these techniques had been assessed in terms of bias and precision, as suggested by the ISMRM quantitative MR study group. We propose a two-stages framework denoted Variable Echo Times neural Network (VET-Net), which considers multi-echo MR images and their echo times to estimate PDFF. VET-Net showed a bias of -1.35% when tested over a multi-site phantom dataset, and a within-standard deviation of 0.81% over liver MR images with different TEs. |
| 5180 | Computer 82
|
Automatic Aorta Quantification and Aneurysm Detection using 3D Deep Learning on Large-scale Non-Cardiac-Gated MRI |
| Thanh-Duc Nguyen1, Saurabh Garg1, Nasrin Akbari1, Saqib Basar1, Sean London2, Yosef Chodakiewitz2, Rajpaul Attariwala1,2, and Sam Hashemi1,2 | ||
1Voxelwise Imaging Technology Inc., Vancouver, BC, Canada, 2Prenuvo, Vancouver, BC, Canada |
||
Keywords: Machine Learning/Artificial Intelligence, Cardiovascular, Aorta quantification We present a validated AI technique for quantification of whole aorta diameter from non-cardiac-gated MRI. The mean absolute error between AI-predicted and radiologist-reported diameter is less than 0.27 and 0.249 cm, while Pearson correlation is 0.72 for the ascending (p-value=0.0002) and 0.74 for the descending aorta (p-values=0.0012). SVM classifiers indicated that a 3.15 cm (AUC=0.93) threshold is used to detect a descending aortic aneurysm while a 4.3 cm (AUC=0.92) threshold is used for ascending aortic aneurysm. Large-scale analysis on 3485 individuals showed aortic diameter to increase with age in male and female populations. |
| 5181 | Computer 83
|
Training Strategies for Convolutional Neural Networks in Prostate T2 Relaxometry |
| Patrick Bolan1, Sara Saunders1, Mitchell Gross1, Kendrick Kay1, Mehmet Akcakaya2, and Gregory Metzger1 | ||
1Center for MR Research / Radiology, University of Minnesota, Minneapolis, MN, United States, 2Electrical and Computer Engineering, University of Minnesota, Minneapolis, MN, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Prostate, Relaxometry This work uses convolutional neural networks (CNNS) with two training strategies for estimating quantitative T2 values from prostate relaxometry measurements, and compares the results to conventional non-linear least squares fitting. The CNN trained with synthetic data in a supervised manner gave lower median errors and better noise robustness than either NLLS fitting or a CNN trained on in vivo data with a self-supervised loss. |
| 5182 | Computer 84
|
Single-shot water/fat separation T2* mapping with multiple overlapping-echo detachment imaging |
| Qing Lin1, Weikun Chen1, Jian Wu1, Taishan Kang2, Xinran Chen1, Zhigang Wu3, Shuhui Cai1, and Congbo Cai1 | ||
1Department of Electronic Science, Xiamen University, Xiamen, China, 2Magnetic Resonance Center, Zhongshan Hospital Afflicated to Xiamen University, Xiamen, China, 3MSC Clinical & Technical Solutions, Philips Healthcare, Shenzhen, China |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Imaging, water/fat separation Most of the water/fat separation techniques need to acquire multiple images with different echo time, which usually take long acquisition time. Multiple overlapping-echo detachment (MOLED) imaging can shorten acquisition time by acquiring multiple MR echo signals in the same k-space. Here, a new method for fast water/fat separation T2* mapping with MOLED technology was proposed, which can obtain quantitative T2* maps and M0 maps of water and fat in a single shot. In vivo experiment demonstrates that the proposed method can obtain accurate quantitative parameter values and accurate separation of water and fat images. |
| 5183 | Computer 85
|
Deep Learning Quantification of Magnetic Resonance Spectroscopy Based on Basis set and Exponential Priors |
| Dicheng Chen1, Huiting Liu1, Yirong Zhou1, Xi Chen2, Zhangren Tu1, Liangjie Lin3, Zhigang Wu3, Jiazheng Wang3, Di Guo4, Jianzhong Lin5, and Xiaobo Qu1 | ||
1Department of Electronic Science, National Institute for Data Science in Health and Medicine, Xiamen University, Xiamen, China, 2McLean Hospital, Harvard Medical School, Belmont, MA, United States, 3Philips Healthcare, Beijing, China, 4School of Computer and Information Engineering, Xiamen University of Technology, Xiamen, China, 5Department of Radiology, The Zhongshan Hospital affiliated to Xiamen University, Xiamen, China |
||
Keywords: Machine Learning/Artificial Intelligence, Brain, Magnetic Resonance Spectroscopy, Quantification Quantification of 1H-MRS is difficult because of the overlapping of individual metabolite signals, non-ideal acquisition conditions, and strong background signal interference. We introduced Deep Learning (DL) method to learn these effects to improve the accuracy of the quantification. Results indicate that, compared with the conventional method LCModel, the proposed Qnet (Quantification deep learning network) shows better quantification for both simulated and in vivo acquired MRS data with lower fitting errors and enhanced stability. |
| 5184 | Computer 86
|
Quantitative MRI Parameter Estimation using Neural Controlled Differential Equations: Proof-of-Concept in Intra-voxel Incoherent Motion |
| Daan Kuppens1, Daisy van den Berg1, Sebastiano Barbieri2, Aart J. Nederveen1, and Oliver J. Gurney-Champion1 | ||
1Radiology & Nuclear Medicine, Amsterdam UMC, Amsterdam, Netherlands, 2Centre for Big Data Research in Health, University of New South Wales Sydney, Sydney, Australia |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Imaging In quantitative MRI, tissue properties are estimated from MRI data using bio-physical models that relate the MRI signal to the underlying tissue properties via model parameters. Deep learning can improve parameter estimation, but is conventionally dependent on the input being either a fixed set of input signals or a series of regularly sampled signals. Neural controlled differential equations (NCDEs) are models that are independent of the configuration of input data. NCDEs have similar performance to state-of-the-art acquisition-specific deep learning methods in estimating intra-voxel incoherent motion parameters. Therefore, NCDEs are a generic purpose tool for parameter estimation in quantitative MRI. |
| 5185 | Computer 87
|
Deep-learning and feature selection for fast, quantitative and specific CEST imaging |
| Cecile Maguin1 and Julien Flament1 | ||
1Université Paris-Saclay, CEA, CNRS, MIRCen, Laboratoire des Maladies Neurodégénératives, Fontenay-aux-roses, France |
||
Keywords: Machine Learning/Artificial Intelligence, CEST & MT, Feature selection We propose an artificial neural network combined with a feature selection scheme for fast, quantitative CEST imaging, designed for specificity. Our NN was evaluated on glucose phantoms and glutamate/glucose mixed phantoms and goes beyond performances of classical fittings approaches. |
| 5186 | Computer 88
|
DeepMRS-Net: QUANTIFICATION OF MAGNETIC RESONANCE SPECTROSCOPY MEGA-PRESS DATA USING DEEP LEARNING |
| Christopher Jiaming Wu1 and Jia Guo2 | ||
1Biomedical Engineering, Columbia University, New York, NY, United States, 2Columbia University, New York, NY, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Brain, Convolutional Neural Networks, Multi-class Regression, Unsupervised Learning Quantification of metabolites in the human brain in vivo from magnetic resonance spectra (MRS) has many applications in medicine and psychology, but it remains a challenging task despite considerable research efforts. In this paper, we propose quantification of metabolites from MEGA-PRESS data using deep learning through an unsupervised learning approach. A regression framework based on the Convolutional Neural Networks (CNN) is introduced for estimation of spectral parameters including the relative concentrations of metabolites, line-broadening, and zero-order phase. The results show that the model is capable of reliably fitting in vivo data. |
| 5187 | Computer 89
|
Myocardial strain generation from cine MR images using an automated deep learning network |
| Dayeong An1 and El-Sayed Ibrahim1 | ||
1Biomedical Engineering, Medical College of Wisconsin, Milwaukee, WI, United States |
||
Keywords: Machine Learning/Artificial Intelligence, Heart Current gold-standard method for obtaining myocardial strain is based on MRI tagged images, although this increases the MRI exam time and requires special analysis software. We propose to use cine MR images to train a deep neural-network to generate myocardial strain based on target strain maps generated from tagged images acquired at the same locations and timepoints as the cine images. The results showed high agreement between the output and target strain maps. Our method not only saves MRI scan time by acquiring only cine images but also pre and post image processing time by quantifying myocardial strains automatically. |
| 5188 | Computer 90
|
Comparison of activation functions for optimizing deep learning models solving QSM-based dipole inversion |
| Simon Graf1, Nora Küchler1, Walter Wohlgemuth1, and Andreas Deistung1 | ||
1University Hospital Halle (Saale), Halle (Saale), Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Susceptibility mapping Deploying deep learning models for quantitative susceptibility mapping is driven by optimizing hyper-parameters and using suitable architectures. We investigated the impact of activation functions on network model training. ELU-, leaky ReLU- and ReLU-models with 16 and 32 initial channels were tested for solving dipole inversion on synthetic susceptibility data. All models showed convergence after completing 100 training epochs. However, the 16-channel-ELU-model achieved low losses after only 20 training epochs and showed similar reconstruction performance to the 32-channel-ELU-model. Using the ELU activation allows the use of smaller network models resulting in fewer memory requirements and less training time. |
| 5189 | Computer 91
|
8-minute Rapid Whole-brain Diffusion Spectral Imaging with Deep Learning-based Reconstruction: A Feasibility Study |
| Yuhui Xiong1, Xiaocheng Wei1, Jiankun Dai1, Yang Fan1, Jie Lu2, and Bing Wu1 | ||
1GE Healthcare MR Research, Beijing, China, 2Department of Radiology, Xuanwu Hospital, Capital Medical University, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence This study aims to shorten the diffusion spectral imaging (DSI) scan time to a clinically acceptable level while providing satisfactory complex white matter fiber structure description as well as accurate diffusion metric quantification by applying deep learning-based reconstruction. Images were acquired using conventional (≈30 min) and rapid (≈8 min) DSI sequences, and reconstructed using conventional and DL-based methods, respectively. Atlas-based fiber-tracking and diffusion metrics quantification from various advanced models were conducted. The results demonstrated that the 8-minute rapid DSI sequence combined with DL-recon can provide complex fiber structure tractography and advanced diffusion metric quantification of satisfactory quality. |
| 5190 | Computer 92
|
Comparison of R1, R2* and PDFF mapping by simultaneous multi-relaxation-time Imaging (TXI) method in hepatic disease |
| Yishuang Wang1, Tianxiang Huang1, Meining Chen2, XU YAN2, and Longlin Yin1 | ||
1Radiology, Sichuan Provincial People's Hospital, Chengdu, China, 2MR Scientific Marketing, Siemens Healthcare, Shanghai, China |
||
Keywords: Machine Learning/Artificial Intelligence, Liver MR multiparametric quantitative techniques have an important role in liver-related diseases.We achieved simultaneous proton density fat fraction(PDFF), R2*and R1 quantification of different liver diseases using a simultaneous multi-relaxation-time Imaging(TXI) technique.We observed differences of PDFF,R2* and R1 mapping between healthy volunteers and patients with cirrhosis,HCC,metastatic carcinoma of liver,and liver transplantation.Rapid scanning increases usage of this method in the clinic.These quantitative values can be biomarkers for assessing liver function and liver tissue characterization. |
| 5191 | Computer 93
|
Quantitative DCE-MRI parameter estimation using Deep Learning Framework in the Brain Tumor Patients |
| Piyush Kumar Prajapati1, Rakesh Kumar Gupta2, and Anup Singh1,3,4 | ||
1Centre for Biomedical Engineering, Indian Institute of Technology Delhi, New Delhi, India, 2Department of Radiology, Fortis Memorial Research Institute, Gurugram, India, 3Yardi School of Artificial Intelligence, Indian Institute of Technology Delhi, New Delhi, India, 4Department of Biomedical Engineering, All India Institute of Medical Sciences, New Delhi, India |
||
Keywords: Machine Learning/Artificial Intelligence, Brain Tracer Kinetic (TK) parametric maps are obtained from Dynamic Contrast Enhanced (DCE) - MRI which aid in the detection and grading of brain tumors. Conventionally, TK maps are obtained using Non-Linear-Least-Square (NLLS) fitting approach, which is time consuming and data noise. In the current study, we implemented a deep learning framework whose backbone is attention networks to estimate TK parametric maps. Transfer learning was performed to extend work from synthetic data to high-grade glioma (HGG) patients’ DCE-MRI data to obtain better quality TK maps in lesser time. |
| 5192 | Computer 94
|
Recovery of T2 distribution from quantitative T2-weighted MRI with physically-driven deep learning |
| Hadas Ben-Atya1 and Moti Freiman1 | ||
1Faculty of Biomedical Engineering, Technion, Haifa, Israel |
||
Keywords: Machine Learning/Artificial Intelligence, Relaxometry Recovery of T2 distribution of tissue from MRI data acquired at multiple echo times has the potential to be used as a biomarker for the assessment of various pathologies, including stroke and epilepsy, investigation of neurodegenerative diseases, and tumor characterization. Current deep neural networks (DNN) for T2 distribution recovery are highly sensitive to variations in the acquisition parameters such as different echo times. We present a new physically-driven DNN model that encodes the TE acquisition parameters as part of its architecture. Our model accurately recovers the T2 distribution, regardless of variations in SNR and in the acquisition parameters. |
| 5193 | Computer 95
|
Constraint function for IVIM quantification using unsupervised learning |
| Wonil Lee1, Beomgu Kang1, Jongyeon Lee1, Georges El Fakhri2, Chao Ma2, Yeji Han3, Jun-Young Chung4, Young Noh5, and HyunWook Park1 | ||
1The School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), Yuseong-gu, Korea, Republic of, 2Massachusetts General Hospital, Boston, MA, United States, 3Department of Biomedical Engineering, Gachon University, Incheon, Korea, Republic of, 4Department of Neuroscience, College of Medicine, Gachon University, Incheon, Korea, Republic of, 5Department of Neurology, Gil Medical Center, Gachon University College of Medicine, Incheon, Korea, Republic of |
||
Keywords: Machine Learning/Artificial Intelligence, Quantitative Imaging, Intravoxel incoherent motion Recently, various methods have been proposed to quantify intravoxel incoherent motion parameters. Many studies have shown that quantification methods using deep learning can accurately estimate IVIM parameters. Unsupervised learning is useful when quantifying IVIM parameters for in-vivo data because it does not require label data. However, in some cases, loss function does not converge as iteration increases. Constraint functions can be used to solve these problems by limiting the range of estimated outputs. In this study, we investigated the effects of constraint function to limit the range of estimated output. |
| 5194 | Computer 96
|
Evaluating different k-space undersampling schemes with iterative and deep learning image reconstruction for fast multi-parameter mapping |
| Kornelius Podranski1, Kerrin J. Pine1, Timoteo Colnaghi2, Andreas Marek2, Patrick Scheibe1, Nico Scherf1,3, and Nikolaus Weiskopf1,4 | ||
1Neurophysics, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 2Max Planck Computing and Data Facility, Garching (Munich), Germany, 3Neural Data Science and Statistical Computing, Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig, Germany, 4Felix Bloch Institute for Solid State Physics, Faculty of Physics and Earth Sciences, Leipzig University, Leipzig, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction, Brain Approaches for accelerating multi-echo gradient echo (ME-GRE) acquisitions as a basis for multi-parameter mapping (MPM) were explored. Fully sampled ME-GRE data were retrospectively undersampled to equispaced Cartesian, CAIPIRINHA and Poisson disc patterns. Echoes were jointly reconstructed with the iterative ENLIVE algorithm and the machine learning/artificial intelligence adapted DeepcomplexMRI (DCMRI) approach. The approaches result in comparable peak signal-to-noise ratio (PSNR) and structural similarity index measure (SSIM), but show different types and different levels of artifacts. The DCMRI approach promises fast reconstruction and flexibility in the choice of undersampling patterns for ME-GRE imaging in the future. |
| 5195 | Computer 97
|
Joint Distribution Modeling for Accelerated T1rho Reconstruction |
| Congcong Liu1, Zhuo-Xu Cui1, Yuanyuan Liu1, Chentao Cao1, Jing Cheng1, Yanjie Zhu1, Haifeng Wang1, and Dong Liang1,2 | ||
1Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China, 2Pazhou Lab, Guangzhou, China |
||
Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction Traditional hand-craft designed methods to accelerated T1rho mapping have limited characterisation capabilities, while deep learning methods lack the interpretability. On the other hand, the joint distribution is the most direct and accurate way to characterize the correlation between different images. Therefore, we attempt to propose a joint distribution estimation method and use it to construct a T1$$$\rho$$$ reconstruction model. In particular, we use a score-based diffusion model to model the joint distribution of acquired T1rho-weighted images. Moreover, the corresponding reconstruction model is solved using the Langevin gradient descent method. Finally, numerical experiments validate the effectiveness of the proposed method. |
| 5196 | Computer 98
|
A multicenter external validation study of the role of AI algorithm in detecting and localizing clinically significant prostate cancer on mpMRI |
| Zhaonan Sun1, Xiaoying Wang2, and Kexin Wang3 | ||
1Department of Radiology, Peking University First Hospital, Beijing, China, 2Department of Radiology, Peking University First Hospital, Beijing, China, 3School of Basic Medical Sciences, Capital Medical University, Beijing, China |
||
Keywords: Machine Learning/Artificial Intelligence, Prostate A total of 557 mpMRI data were retrospectively collected from three hospitals to build an external validation dataset, with 245 csPCa cases and 312 non-csPCa cases. The csPCa lesions were annotated based on pathology records by two experienced radiologists. AI algorithms were used to automatically detect and localize the suspicious csPCa areas on the T2WI and ADC maps. The metrics of sensitivity, specificity, and accuracy were used to evaluate the diagnostic efficacy of the AI algorithms at the lesion level, the sextant level, and the patient level. |
| 5197 | Computer 99
|
Glioma Classifications with 7T MR Spectroscopic Imaging |
| Sukrit Sharma1, Cornelius Cadrien2, Philipp Lazen1, Hangel Gilbert1, Roxane Licandro3, Wolfgang Bogner1, and Georg Widhalm2 | ||
1High-field MR Center, Department of Biomedical Imaging and Image-guided Therapy, Medical University of Vienna, Vienna, Austria, Medical University of Vienna, Vienna, Austria, 2Medical University of Vienna, Vienna, Austria, 3Department of Biomedical Imaging and Image-guided Therapy, Computational Imaging Research Lab (CIR) and Laboratory for Computational Neuroimaging, A.A. Martinos Center for Biomedical Imaging, Massachusetts General Hospital / Harvard Medical School, Charlestown, MA, US., Medical University of Vienna, Vienna, Austria |
||
Keywords: Machine Learning/Artificial Intelligence, Brain, Glioma To contribute to better tumour classification and thus enhancing patient outcomes, we statistically analysed metabolic maps of 37 glioma patients obtained using high resolution 7T MRSI. We tested and optimised different semi-supervised learning based classification approaches. Random forest classification of IDH mutation status and tumour grade in clinical imaging based segmented tumour regions yielded high diagnostic accuracy with AUC of 86% and 99% respectively. We found Glu, Gln, GSH, tCho, Ins, Gly and tCr as important determining features. These are similar to comparable SVS studies while providing the advantage of whole-brain coverage. |
| 5198 | Computer 100
|
3D CNN for Oxygen Extraction Fraction Mapping with combined QSM and qBOLD |
| Patrick Kinz1 and Lothar R Schad1 | ||
1Computer Assisted Clinical Medicine, Medical Faculty Mannheim, Heidelberg University, Mannheim, Germany |
||
Keywords: Machine Learning/Artificial Intelligence, Oxygenation We developed a CNN for OEF mapping from QSM+qBOLD data, which utilizes utilizes 3D convolutional layers. Two dimensions for the spatial components of an image and one dimension for the temporal component in the qBOLD data. The results are an improvement over our previous 2D CNN, but even the more advanced network architecture struggles with voxels, that have a very low deoxyhemoglobin content. In this abstract we also study with simulated data when the CNN produces reliable results and when it predicts default values for the reconstructed parameters instead. |