To begin searching the abstracts, please use the search feature above.

Digital Poster - Musculoskeletal Weekend and Oral

Digital Poster (no CME credit)

SMRT Ed Session

SMRT Poster Presentations (no CME credit)

Acq/Recon/Analysis Digital Poster	Body Digital Poster	Cancer/Spectroscopy/Molecular Imaging/Pre-Clinical Digital Poster	Cardiovascular Digital Poster
Contrast Mechanisms Digital Poster	COVID-19 Digital Poster	Diffusion/Perfusion Digital Poster	Engineering/Interventional/Safety Digital Poster
fMRI Digital Poster	MR Value Digital Poster	Neuro Digital Poster	Pediatrics Digital Poster

Musculoskeletal Digital Poster (No CME Credit)

Parent Session Title

Bone, Cartilage & Joint MRI

Program # 2946 - 2985

Muscle MRI

Program # 3165 - 3204

Artificial Intelligence Applied to MSK MRI

Program # 4045 - 4084

Quantitative MSK MRI

Program # 4203 - 4241

Cartilage I

Bone, Cartilage & Joint MRI

Musculoskeletal

2946		Relaxation Time Mapping of Knee Articular Cartilage Using Magnetic Resonance Fingerprinting
		Sanam Assili¹, Victor Casula^1,2, Jaakko Ikäheimo¹, Egor Panfilov¹, Ari Väärälä¹, Martijn A. Cloos³, Riccardo Lattanzi⁴, and Miika T. Nieminen^1,2,5
		¹Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland, ²Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland, ³Centre for Advanced Imaging, University of Queensland, Brisbane, Queensland, Australia, ⁴Center for Advanced Imaging Innovation and Research, New York University Grossman School of Medicine, New York, NY, United States, ⁵Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland
		Knee osteoarthritis is a degenerative musculoskeletal disorder which is characterized by progressive damage of articular cartilage (AC). Quantitative MRI (qMRI), such as T1 and T2 maps, can be used to assess the AC compositional changes. However, long acquisition time and equipment dependency are the limitations of the current clinical qMRI techniques. In the present study, we validate magnetic resonance fingerprinting (MRF) against conventional T1 and T2 relaxation time mapping methods in cartilage-mimicking phantoms, and apply MRF in vivo for sub-regional analysis of femoral and tibial cartilage in human subjects.

2947		Deriving a Cartilage Signature to Predict Joint Replacement in Osteoarthritis
		Edward Peake^1,2,3, Stefan Pszczolkowski^1,2,3, Christoph Arthofer^2,3,4, and Dorothee P Auer^1,2,3
		¹NIHR Nottingham Biomedical Research Centre, University of Nottingham, Nottingham, United Kingdom, ²Radiological Sciences, Division of Clinical Neuroscience, University of Nottingham, Nottingham, United Kingdom, ³Sir Peter Mansfield Imaging Centre, University of Nottingham, Nottingham, United Kingdom, ⁴Nuffield Department of Clinical Neuroscience, University of Oxford, Oxford, United Kingdom
		Prediction of future knee replacement in osteoarthritis may support tailored decision making, and provide a surrogate outcome marker for clinical trials of diseases modifying osteoarthritis drugs. Fully automated cartilage segmentation is generated from knee MRI, and feature are extraction to create a radiomic signature for the prediction of knee replacement within 5 -years. The radiomic signature using univariate cox regression predicted surgery with a HR of 7.5 (p = 2e-28, 95% CI 7.1 – 7.9) which was higher than established clinico-demographic multivariate prediction model with HR 5.9 (p = 2e-28, 95%CI 5.6 – 6.2).

2948		T1ρ relaxation times for voxel-wise characterization of longitudinal changes in hip cartilage biochemistry
		Koren Roach¹, Richard Souza¹, Sharmila Majumdar¹, and Valentina Pedoia¹
		¹UCSF, San Francisco, CA, United States
		Hip osteoarthritis coincides with or is preceded by biochemical changes in articular cartilage,^{1, 2} which can be measured on a voxel-wise basis with magnetic resonance T_1ρ and T₂ relaxation times. In this study, we employed principal component analysis to characterize variation in longitudinal changes of hip cartilage T_1ρ relaxation times across healthy subjects and those with early-to-moderate hip osteoarthritis. Our results indicated that over two years, a majority of the variation in T_1ρ relaxation time changes was located in the deep layers of the acetabular cartilage.

2949		The value of T2 mapping to characterize early knee joint cartilage damage in hemophilia arthropathy
		Shufang Wei¹, Jiajia Li¹, Xianchang Zhang², Jing An³, and Yinghui Ge¹
		¹Fuwai Central China Cardiovascular Hospital, Zhengzhou, China, ²MR Collaboration, Siemens Healthcare Ltd, Beijing, China, ³Siemens Shenzhen Magnetic Resonance Ltd, Shenzhen, China
		Detecting early articular cartilage damage can help formulate preventive treatment plans for hemophilia arthropathy (HA). Magnetic responance imaging (MRI) T2 mapping has been used extensively to investigate early chondral changes in osteoarthritis; however, it has seldomly been applied to investigate HA. This study acquired T2-mapping results on 30 patients (54 knees) with HA and 15 volunteers (24 knees). Patients with HA had significantly higher knee cartilage T2 values than the healthy controls. The T2 values also increased with the disease severity, suggesting that MRI T2 mapping is a promising method for detecting early stage cartilage damage in patients with hemophilia.

2950		Optimization of Relaxation along Fictitious Field (RAFF) Contrast to Detect Osteoarthritis
		Seyed Amir Mirmojarabian¹, Victor Casula¹, Olli-Pekka Aro¹, Henning Henschel², Miika T Nieminen^1,3,4, and Timo Liimatainen^1,3
		¹Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland, ²Department of Medicinal Chemistry, Uppsala University, uppsala, Sweden, ³Medical Research Center, Oulu University Hospital, Oulu, Finland, ⁴Department of Diagnostic Radiology,, Oulu University Hospital, Oulu, Finland
		Relaxation along Fictitious Field (RAFF) at various pulse durations was exploited to study the contrast in collagen phantoms and cartilage-mimicking phantoms. RAFF Relaxation times from collagen phantoms agreed with Bloch McConnell simulations. The maximum contrast between collagen phantoms with concentration difference was maximized when pulse duration was ~2.8 ms. Given the optimized pulse duration, our simulation results combined with ex vivo measurements suggest that hydroxyl groups concentration through proton exchange between collagen and free water may have significant role in T_RAFF contrast between intact and degraded cartilage.

2951		The value of magnetic resonance T2*mapping for early detection of knee cartilage damage in hemophilia arthropathy
		Jiajia Li¹, Shufang Wei¹, Xianchang Zhang², Jing An³, and Yinghui Ge¹
		¹Fuwai Central China Cardiovascular Hospital, Zhengzhou, China, ²MR Collaboration, Siemens Healthcare Ltd., Beijing, China, Beijing, China, ³Siemens Shenzhen Magnetic Resonance Ltd, Shenzhen, China
		Early detection and therapy of articular cartilage damage are key to the management of hemophilia arthropathy (HA). Magnetic resonance (MR) T2* mapping has been used to detect iron deposits in the brain and liver but rarely in HA. We compared T2mapping results of 15 HA patients (18 knees) and 10 volunteers (18 knees). We found that the average T2 values of the medial femoral condyle and medial tibial plateau cartilage in HA patients were significantly lower than in the normal group, suggesting the importance of T2* mapping in the quantitative assessment of iron deposition for early cartilage damage diagnosis in HA patients.

2952		Decreased collagen content in tendons of patients with osteoporosis and osteopenia detected with ultrashort echo time cones MRI
		Saeed Jerban¹, Yajun Ma¹, Amir Masoud Afsahi¹, Douglas G Chang², Zhao Wei¹, Meghan Shen¹, Mei Wu¹, Alecio Lombardi^1,3, Nicole Le⁴, Jiang Du¹, and Eric Y Chang^1,3
		¹Radiology, University of California, San Digeo, La Jolla, CA, United States, ²Orthopaedic Surgery, University of California, San Digeo, La Jolla, CA, United States, ³Radiology Service, VA San Diego Healthcare System, San Diego, CA, United States, ⁴Radiology, VA San Diego Healthcare System, La Jolla, CA, United States
		Bone and tendon comprise a highly interactive mechanical unit. Investigating tendons quality during osteoporosis and osteopenia progress is of great interests. Clinical MRI sequences are not often capable of directly visualizing tendon because of the tissue’s short T2. Ultrashort echo time (UTE) MRI combined with magnetization transfer (MT) modeling (UTE-MT) has demonstrated promise as a quantitative technique that is resistant to the magic angle effect. Lower leg tendons in osteopenia and osteoporosis patients were compared with healthy subjects via UTE-MT. Macromolecular fraction (MMF) obtained from UTE-MT modeling showed a significant reduction in osteopenia and osteoporosis patients compared with healthy subjects.

2953		Highly Efficient Single-Point Dixon-Based Fat Suppression for Ultrashort Echo Time Double Echo Steady State (UTE-DESS) Imaging of the Knee Joint
		Hyungseok Jang¹, Yajun Ma¹, Michael Carl², Saeed Jerban¹, Eric Y Chang^1,3, and Jiang Du¹
		¹University of California, San Diego, San Diego, CA, United States, ²GE Healthcare, San Diego, CA, United States, ³Veterans Affairs San Diego Healthcare System, San Diego, CA, United States
		In this study, we explored the feasibility of applying single-point Dixon (spDixon)-based fat suppression to a novel ultrashort echo time-based double echo steady state (UTE-DESS) sequence for highly efficient morphological musculoskeletal (MSK) imaging. Compared to the two-point Dixon-based approach used with conventional UTE-DESS MSK imaging which requires two echoes for fat-water separation, we demonstrated a new technique which requires only a single image in, removing the need for additional data acquisitions. The feasibility and efficacy of our approach were demonstrated in human knee joints.

2954		Examining short-term longitudinal and activity-based variability of femoral cartilage T2 relaxation times in healthy subjects
		Lauren Watkins^1,2, Andrew Schmidt¹, Elka Rubin¹, Marco Barbieri¹, Arjun Desai^1,3, Valentina Mazzoli¹, Marianne Black¹, Garry Gold^1,2, Brian Hargreaves^1,2,3, Akshay Chaudhari^1,4, and Feliks Kogan¹
		¹Radiology, Stanford University, Stanford, CA, United States, ²Bioengineering, Stanford University, Stanford, CA, United States, ³Electrical Engineering, Stanford University, Stanford, CA, United States, ⁴Biomedical Data Science, Stanford University, Stanford, CA, United States
		Subtle variations in T2 relaxation times may be related to the risk of early osteoarthritis and progression. It is important to estimate T2 variability in healthy individuals as a basis for cross-sectional and longitudinal comparisons. We examine short-term T2 variability in femoral articular cartilage of healthy volunteers over 5 days and 5 weeks and consider the impacts of physical activity. Over both periods, mean T2 change, variance, and CV_RMS were small and comparable between knees and cartilage subregions. There was a moderate positive correlation between T2 changes and weekly rated perceived exertion over 5 weeks

2955		Morphologic Structure of Rabbit Suprapatella Cartilage by µMRI and PLM
		Hannah Mantebea¹, Syeda Batool¹, Mouhamad Hammami¹, and Yang Xia¹
		¹Physics, Oakland University, Rochester, MI, United States
		This study aims to quantity the structural morphology of sesamoid fibrocartilage in rabbits, using µMRI at 19.5-µm resolution and polarized light microscopy at 1-µm resolution. We show that the sesamoid fibrocartilage has one thin surface layer (~ 10 µm) of parallel-arranged fibers followed immediately by the majority of random fibers (~ 390 µm). T2 anisotropy was not observed in fibrocartilage. The collagen fibers in the surface layer are dense and in parallel with the surface, which allows it to withstand stress from the quadriceps tendon during flexion, while at the same time to reduce pressure load from the femur.

2956		Deep Cartilage UTE-T2* Shows Compositional Heterogeneity in Patients with Degenerative Meniscus Tears
		Ashley A. Williams^1,2, Karyn E. Chappell^1,2, and Constance R. Chu^1,2
		¹Orthopaedic Surgery, Stanford University, Stanford, CA, United States, ²Veterans Affairs Palo Alto Health Care System, Palo Alto, CA, United States
		This study evaluated deep cartilage and meniscus UTE-T2* against spatially targeted arthroscopic assessments in patients with degenerative meniscus tears to further inform interpretation and utility of in vivo UTE-T2* for different degrees of pathology. While arthroscopic evaluation confirmed that UTE-T2* was elevated in menisci with degenerative fraying and tears, increasing intra-operative cartilage grade was not associated with strictly increasing cartilage UTE-T2. Instead, in osteoarthritic disease states where the articular surface of cartilage is disrupted, this report suggests that mono-exponential deep cartilage UTE-T2 may not be a reliable indicator of tissue pathology.

2957		Evaluation of Lesion Tissue Composition in Patients with Juvenile Osteochondritis Dissecans (JOCD) of the Knee Using T2* Mapping at 3T
		Stefan Zbyn^1,2, Cassiano Santiago¹, Casey P. Johnson^1,3, Kai D. Ludwig^1,2, Lin Zhang⁴, Shelly Marette², Marc A. Tompkins⁵, Bradley J. Nelson⁵, Takashi Takahashi², Gregory J. Metzger¹, Cathy S. Carlson³, and Jutta M. Ellermann^1,2
		¹Center for Magnetic Resonance Research, University of Minnesota, Minneapolis, MN, United States, ²Department of Radiology, University of Minnesota, Minneapolis, MN, United States, ³Department of Veterinary Clinical Sciences, University of Minnesota, St. Paul, MN, United States, ⁴Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN, United States, ⁵Department of Orthopaedic Surgery, University of Minnesota, Minneapolis, MN, United States
		Clinical MRI of patients with juvenile osteochondritis dissecans (JOCD) relies on morphological, spin echo imaging with long echo times that cannot unambiguously distinguish between the cartilaginous, fibrous and osseous tissues within lesions. This 3T study evaluates 34 JOCD lesions from 25 patients using morphological MRI and T₂* mapping. JOCD progression was associated with a significant T₂* decrease in progeny and interface. Additionally, significantly lower T₂* in stage IV progeny and parent bone of stable lesions suggests increased bone density compared to the healthy, control bone. T₂* mapping provides quantitative evidence of disease progression which may help inform clinical JOCD management.

2958		Disruption of collagen fiber architecture after DMM surgery using high-resolution DTI
		Nian Wang¹, Anthony J. Mirando², Yi Qi³, Matthew J. Hilton², and Charles E. Spritzer³
		¹Radiology and Imaging Sciences, Indiana University, Indianapolis, IN, United States, ²Department of Orthopaedic Surgery, Duke University, Durham, NC, United States, ³Department of Radiology, Duke University, Durham, NC, United States
		Nondestructively probing the 3D collagen fiber architecture of articular cartilage is still challenging for both clinical and preclinical studies. Diffusion tensor imaging (DTI) is able to resolve the collagen fiber directions by quantifying the unique water diffusion properties in soft tissues. Applying DTI and tractography in cartilage is limited but may provide noninvasive imaging biomarkers for Osteoarthritis (OA). In this study we imaged rat knees after 8 weeks destabilization of the medial meniscus (DMM) surgery using a modified 3D diffusion-weighted spin-echo pulse sequence at both high spatial resolution and high angular resolution.

2959		Learned knee cartilage and meniscus shape features are associated with osteoarthritis incidence
		Claudia Iriondo¹, Jinhee Lee¹, Sharmila Majumdar¹, and Valentina Pedoia¹
		¹University of California, San Francisco, San Francisco, CA, United States
		We derive cartilage and meniscus point clouds from 40,796 high resolution knee MR images and train point cloud networks to extract osteoarthritis shape features. We demonstrate the utility of these learned features by assessing their relative contributions in a Cox Proportional Hazard Regression model with existing clinical risk factors predicting incident radiographic osteoarthritis. Shape biomarkers for tibiofemoral joint cartilage and menisci had significantly increased hazard ratios. The best performing shape biomarker– tibial and femoral cartilage shape– combined with clinical risk factors achieved a concordance index of 0.759. Our findings suggest point cloud learned shape features are promising OA biomarkers.

2960		Assessment of knee cartilage volume, thickness and T2 relaxation times in patients with osteoarthritis
		Hui Tan¹, Bin Wang², Wulin Kang¹, Qiuju Fan¹, Nan Yu¹, Shaoyu Wang³, Esther Raithel⁴, Yue Li², and Tuona Di²
		¹The affiliated hospital of Shaanxi University of Chinese Medicine, Xianyang, China, ²Shaanxi University of Chinese Medicine, Xianyang, China, ³Siemens Healthineers, Shanghai, China, ⁴Siemens Healthcare, Erlangen, Germany
		This study aimed to quantitative measurements of the cartilage volume , thickness and T2 relaxation times in patients with symptomatic mild to moderate KOA by using T2 mapping in 3 T MR. The prototypic software were used to evaluate 17 mild KOA and 14 moderate KOA patients. Both biochemical imaging and morphologic examinations were sensitive to study the mild to moderate KOA. Compared with the mild, the moderate OA patients exhibited decreased CV and CT with the growth of K-L grade, and the T2 value increased with the growth of K-L grade.

2961		Imaging hyaluronan-mediated inflammation in articular cartilage
		Jose M Raya¹, Alejandra M Duarte¹, Dalibel M Bravo¹, Elisa M Ramos¹, Chongda M Zahng¹, Mary M Cowman², Thorsten M Kirsch¹, Mark Wilne³, Len Lyut³, and Amparo M Ruiz¹
		¹New York University Langone Health, New York, NY, United States, ²New York University Tandon School of Engineering, New York, NY, United States, ³University of Western Ontario, Ontario, ON, Canada
		The objective of this work is to develop and validate a molecular imaging approach to image inflammation in articular cartilage. We targeted hyaluronan (HA)-mediated inflammation, which is a common inflammatory pathway activated in cartilage, with contrast agents designed using a 15-mer peptide that modulates HA-cell interactions. We identify optimal imaging times and demonstrate that contrast agent accumulates in areas of macroscopic damage. In vivo, MRI detected accumulation of the contrast agent in cartilage of PTOA joints but not in controls. Cartilage in all but one ACL-injured joint showed T₁-changes indicative of inflammation, demonstrating that HA-mediated inflammation is a common finding.

2962		Improving Fast 3D-T1rho Mapping of Human Knee Cartilage with Data-Driven Learned Sampling Pattern
		Marcelo Victor Wust Zibetti¹, Azadeh Sharafi¹, and Ravinder Regatte¹
		¹Radiology, NYU Langone Health, New York, NY, United States
		in evaluating the performance of optimized SP for accelerating T1rho mapping of the knee cartilage. In this study, we investigate the improvements in accelerating the T1rho mapping of knee joint by learning the SP in a data-driven manner. It was observed that the optimal learned SP depends on the selected spatial-temporal (k-t) data and the chosen reconstruction. Our preliminary results show that the learned SP improved the quality of the accelerated T1rho mapping of knee cartilage over Poisson disk for several different kinds of CS reconstructions.

2963		Application value of 3D-MRI based on compressed SENSE technology in meniscal injuries
		Peiqi Ma¹, Yushan Yuan¹, Zongxi Zhang¹, Bin Peng¹, Jian Xu¹, and Xiuzheng Yue²
		¹Fuyang People's Hospital，Anhui Province，China, Fuyang, China, ²Philips Healthcare, Beijing, China, Beijing, China
		The compressed SENSE(CS-SENSE) technique, which is the combination of the parallel imaging technique SENSE with compressed sensing, has now been widely used in acceleration of MR examinations of various anatomies and anatomical contrasts. The purpose of this study was to explore the possibility of CS-SENSE in 3D knee joint imaging and its diagnostic value for meniscus injury types.

2964		Regional Variations in T2 Relaxations Times in the Hip Cartilage of Female Water Polo Players and Synchronized Swimmers
		Elka B Rubin¹, Joanna L Langner², Marianne S Black², Arjun D Desai², James W MacKay³, Carly Jones⁴, Kimberly E Hall², Marc R Safran², Feliks Kogan², and Garry E Gold²
		¹Radiology, Stanford University, Stanford, CA, United States, ²Stanford University, Stanford, CA, United States, ³Radiology, Cambridge University, Cambridge, United Kingdom, ⁴Centre for Hip Health and Mobility, University of British Columbia, Vancouver, BC, Canada
		Water polo players and synchronized swimmers have been found to have an increased prevalence of femoroacetabular impingement morphology compared to the general population. In this study, we used quantitative MRI to identify regional patterns in the composition of hip cartilage of Division 1 female water polo players and synchronized swimmers. Variations in T₂ relaxation times showed regional hip cartilage differences between the water polo players and synchronized swimmers and within the sub-regions of the synchronized swimmer’s hip cartilage. This study suggests that microstructural differences are present in the various regions of these athletes’ hip cartilage.

2965		Deep, intermediate and superficial layers of patella cartilage assessment using T2 mapping in patients with chondromalacia.
		Elena Voronkova^1,2, Petr Menshchikov^3,4, Ilya Melnikov¹, Andrei Manzhurtsev^1,4, Maxim Ublinskii^1,4, Denis Vorobyev¹, Dmitriy Kupriyanov³, and Tolib Akhadov¹
		¹Clinical and Research Institute of Emergency Pediatric Surgery and Trauma, Moscow, Russian Federation, ²National Research Nuclear University MEPhI, Moscow, Russian Federation, ³Philips Healthcare, Moscow, Russian Federation, ⁴Emanuel Institute of Biochemical Physics of RAS, Moscow, Russian Federation
		In this work, the values of the transverse relaxation times (T2) were measured for the first time in the deep, intermediate, and superficial layers of the patellar cartilage at different stages of chondromalacia as compared to control. The T2 values in the intermediate layer of cartilage show the best sensitivity to the grade of injury and have statistically significant differences for the groups of normal, mild, and severe cartilage damage. The T2-mapping technique is a potential tool for studying structural changes in cartilage and T2 values can be considered as a non-invasive biomarker for the assessment of the cartilage condition.

Cartilage II

Bone, Cartilage & Joint MRI

Musculoskeletal

2966		T2 mapping of normal and abnormal cartilage in the equine distal interphalangeal joint using low field MRI
		Melissa E Baker¹, Lucy E Kershaw², Steve Roberts³, Richard Reardon¹, Sionagh Smith¹, and Sarah E Taylor¹
		¹Royal (Dick) School of Veterinary Studies and The Roslin Institute, The University of Edinburgh, Edinburgh, United Kingdom, ²Centre for Cardiovascular sciences and Edinburgh Imaging, The University of Edinburgh, Edinburgh, United Kingdom, ³Hallmark Veterinary Imaging Ltd, Guildford, United Kingdom
		Osteoarthritis is a common source of pain and lameness in horses of all ages. There are currently no diagnostic tests available to detect early cartilage lesions. A T₂ mapping sequence was used on a low field 0.27 T open MR system to image the cartilage of 20 ex vivo equine distal interphalangeal joints. Results showed significantly increased mean T₂ relaxation time 114±95 ms in lateral cartilage (higher OARSI grades) compared to 71±21 ms in medial cartilage (lower OARSI grades). This quantitative MRI technique could be used to diagnose early pathology in equine joints in the future.

2967		Automated Pipeline for Quantitative MRI Evaluation of Knee Articular Cartilage in Longitudinal Osteoarthritis Trials
		Vladimir Juras¹, Veronika Janáčová¹, Pavol Szomolanyi¹, Markus Schreiner², Didier Laurent³, Celeste Scotti³, and Siegfried Trattnig¹
		¹Department of Biomedical Imaging and Image-Guided Therapy, Medical University of Vienna, Vienna, Austria, ²Department of Orthopedics and Trauma Surgery, Medical University of Vienna, Vienna, Austria, ³Novartis Institutes for Biomedical Research, Basel, Switzerland
		The purpose of this study was to develop a fully automated reproducible tool for analysis of morphological and compositional quantitative MR parameters in well-defined cartilage sub-fields across the human knee. This tool was deployed to proton and sodium MR images and the reproducibility was assessed by test re-test scan of ten patients at baseline and after eight days. Due to high reproducibility found in this study this tool may be considered as a good alternative to manual evaluation which, as being extremely expertise-dependent and time-consuming, could represents a considerable burden for large clinical osteoarthritis trials.

2968		Automated Knee Cartilage Thickness Measurement from Magnetic Resonance Images
		Yongcheng YAO¹, Sheheryar KHAN¹, Junru ZHONG¹, Siyue LI¹, and Weitian CHEN¹
		¹Department of Imaging and Interventional Radiology, The Chinese University of Hong Kong, Hong Kong, China
		We propose an automated method to measure knee cartilage thickness from MR images. The proposed method comprises the processes including cartilage and bone segmentation, cartilage boundary separation, surface normal estimation, and uncertainty estimation in thickness measurement. Compared with the conventional nearest neighbor approach, the proposed method provides more reliable measurement at the edges and the regions with thin cartilage.

2969		Wrist cartilage segmentation using U-Net convolutional neural networks
		Nikita Vladimirov¹ and Ekaterina A. Brui¹
		¹Department of Physics and Engineering, ITMO University, Saint Petersburg, Russian Federation
		Segmentation of wrist cartilage may be of interest for the detection of cartilage loss during osteo- and rheumatoid arthritis. In this work, U-Net convolutional neural networks were used for automatic wrist cartilage segmentation. The networks were trained on a limited amount of labeled data (10 3D VIBE images). The results were compared with the previously published for a planar patch-based archutecture (3D DSC = 0.71). Utilisation of U-Net archutecture and data augmentation allowed to significantly increase the segmentation accuracy in lateral slices. Truncation of the deepest level in the classical U-Net architecture provided the 3D DSC=0.77.

2970		3D Texture Analysis of 3D DESS Cartilage Images for Prediction of Knee Osteoarthritis
		Daniel Uher¹, Ari Väärälä¹, Antti Isosalo¹, Victor Casula^1,2, and Miika T. Nieminen^1,2,3
		¹Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland, ²Medical Research Center, University of Oulu and Oulu University Hospital, Oulu, Finland, ³Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland
		In this study, a gray level co-occurrence matrix (GLCM) based 3D Texture Analysis method was utilized for early prediction of knee osteoarthritis using 3D DESS images. Twenty subjects were extracted from the Osteoarthritis Initiative (baseline) with Kellgren-Lawrence (KL) score = 0 at baseline. Ten of the selected subjects developed the disease and showed KL ≥ 2 at the 36-month visit. Knee DESS images were analyzed using various quantization schemes and three machine learning models were trained based on the output GLCM features. Naïve Bayes model trained on tibial features showed the highest accuracy (86.8%) for OA onset after 36 months.

2971		GagCEST Imaging of Healthy and OA Patients at 7 Tesla
		Blake Alexander Benyard¹, Ryan Armbruster¹, Abigail Cember¹, Neil Wilson², Ravinder Reddy¹, and Joshua Baker³
		¹Radiology, University of Pennsylvania, Philadelphia, PA, United States, ²Siemens Medical Solutions USA Inc, Malvern, PA, United States, ³Penn Medicine, Rheumatology, Philadelphia Veterans’ Affairs Medical Center, Philadelphia, PA, United States
		Imaging of knee cartilage is of enormous significance, particularly for patients with osteoarthritis (OA). In this work, we applied Glycosaminoglycans-weighted Chemical Exchange Saturation Transfer (GagCEST) magnetic resonance imaging (MRI) at 7 Telsa (7T) to OA and healthy patients. GagCEST MRI revealed an increase in GagCEST (%) values for control patients in comparison to OA patients and can serve as a sensitive molecular biomarker for quantifying early metabolic changes in cartilage.

2972		OPTIMIZATION OF KNEE CARTILAGE TEXTURE ANALYSIS FROM QUANTITATIVE MRI T2 MAPS
		Veronika Janáčová¹, Pavol Szomolanyi¹, Siegfried Trattnig¹, and Vladimir Juras¹
		¹High Field MR Centre, Department of Biomedical Imaging and Image-Guided Therapy, Medical University of Vienna, Vienna, Austria
		Texture analysis is currently being explored as an tool for evaluation of collagen specific T2 mapping in magnetic resonance imaging in osteoarthritis (OA) research. We assessed the effect of GLCM calculation offset and number of parameters used for T2 mapping of OA lesion and healthy cartilage on GLCM features. Significant differences in autocorrelation, contrast and homogeneity between lesion and reference when using combination of different offsets (0° and 90°) and 2- and 3-parameter T2 mapping were found. Considering architecture of healthy and OA cartilage and resultant T2 distribution, autocorrelation, contrast and homogeneity might be reliable indicators of cartilage damage.

2973		Ultra high resolution UTE imaging of the knee at 7T: Simultaneous view of cartilage, meniscus, ligament, tendon, and the chondro-osseous junction
		Yongxian Qian¹, Gregory Chang¹, Eric J. Strauss², and Fernando E. Boada¹
		¹Radiology, New York University, New York, NY, United States, ²Orthopaedic Surgery, New York University, New York, NY, United States
		This human subject study presents preliminary results demonstrating the great potential of ultra-high resolution (0.14mm) UTE imaging to simultaneously visualize daily movement active tissues in the knee joint, including cartilages, menisci, ligaments, tendons, and even the chondro-osseous junctions. Non-invasive observation of these functional tissues are critical to understanding mechanisms of the onset and progression of osteoarthritis (OA).

2974		An efficient $$$R_{1\rho}$$$ dispersion imaging method for the human knee cartilage using constant magnetization prepared turbo-FLASH
		Yuxi Pang¹, Riann Palmieri-Smith^2,3, and Tristan Maerz³
		¹Dept. of Radiology, University of Michigan, Ann Arbor, MI, United States, ²School of Kinesiology, University of Michigan, Ann Arbor, MI, United States, ³Dept. of Orthopaedic Surgery, University of Michigan, Ann Arbor, MI, United States
		An efficient $$$R_{1\rho}$$$ dispersion imaging method is presented for clinical studies of the human knee cartilage at 3T. Eight constant magnetizations were prepared by tailoring both duration and amplitude of a fully-refocused spin-lock preparation pulse. The limited initial magnetization dynamic range was expanded by the measure from the magic angle location in the deep femoral cartilage. The proposed method was applied to four asymptomatic knees from three subjects. The results from repeated scans and from comparisons with the literature indicate that the proposed method is a promising tool to further explore $$$R_{1\rho}$$$ dispersion of human knee cartilage in clinical settings.

2975		Effects of Angular Resolution and b Value on Diffusion Tensor Imaging in Knee Joint
		Qi Zhao¹, Rees P. Ridout¹, Jikai Shen¹, and Nian Wang²
		¹Duke University, Durham, NC, United States, ²Radiology and Imaging Sciences, Indiana University, Indianapolis, IN, United States
		Diffusion MRI and tractography in the musculoskeletal system has recently attracted more and more attention to investigate the connective tissue microstructure, local collagen fiber alignment, and the 3D collagen fiber network. Although diffusion tractography was successfully performed in some individual regions of knee, investigating the effects of the acquisition parameters for the whole joint are still lack.

2976		Simultaneous Bilateral T1, T2, and T1ρ Relaxation Mapping of the Hip Joint with Magnetic Resonance Fingerprinting
		Azadeh Sharafi¹, Marcelo V. W. Zibetti¹, Gregory Chang¹, Martijn Cloos², and Ravinder Regatte¹
		¹Radiology, NYU Langone Health, New York, NY, United States, ²University of Queensland, Brisbane, Australia
		MR imaging has an essential role in the diagnosis of hip disorders such as femoroacetabular impingement (FAI) and osteoarthritis (OA). An increase in pre-contrast T1 is reported in association with cartilage edema and fibrillation and both T2 and T1ρ are shown to be elevated in OA and FAI patients . This work investigates the feasibility of using the MRF sequence introduced in for simultaneous bilateral T1, T2, and T1ρ mapping of the hip joint and assessing the difference between the left and right side cartilage subregions.

2977		3D sub-millimeter isotropic knee cartilage T1ρ mapping using multi-interleaved fluid-attenuated TSE acquisition (MIXTURE)
		Shinji Saruya¹, Masashi Suzuki¹, Masami Yoneyama², Kaiji Inoue¹, Eito Kozawa¹, and Mamoru Niitsu¹
		¹Department of Radiology, Saitama Medical University, Saitama, Japan, ²Philips Japan, Tokyo, Japan
		For T1ρ quantification, a three-dimensional (3D) acquisition is desired to obtain high-resolution images. In this study, we introduced a new sequence, termed MIXTURE (Multi-Interleaved X-prepared TSE with inTUitive RElaxometry), based on 3D magnetization-prepared multi-interleaved TSE, to achieve rapid and robust acquisition of 3D T1ρ mapping data. MIXTURE might be useful for improving the quality and efficiency of 3D T1ρ-mapping.

2978		Quantitative Magnetization Transfer in the Knee Meniscus: Minimizing Data Collection Protocols
		Fatemeh Mostafavi¹, Lumeng Cui¹, Brennan Berryman², Ives R. Levesque³, and Emily J. McWalter^1,2
		¹Division of Biomedical Engineering, University of Saskatchewan, Saskatoon, SK, Canada, ²Department of Mechanical Engineering, University of Saskatchewan, Saskatoon, SK, Canada, ³McGill University, Montreal, QC, Canada
		A major challenge in the use of quantitative magnetization transfer (qMT) is the long scan time. One method to reduce scan time is acquiring fewer MT-weighted images. qMT mapping of the meniscus was performed in six cadaver knees. Fitting was repeated after systematically decreasing the number of input scans and comparing the results to a reference dataset (10 MT-weighted images). A set of eight MT-weighted images fit to a two-pool model with Gaussian lineshape was found to have an acceptable level of agreement (a priori of 10%) with the reference dataset, saving approximately 8 minutes in scan time.

2979		A generalized magic angle effect model for better characterizing anisotropic T2W signals of human knee femoral cartilage
		Yuxi Pang¹
		¹Dept. of Radiology, University of Michigan, Ann Arbor, MI, United States
		Collagen fibril microstructural distributions in articular cartilage are extremely complex. The simplistic two halves radially segmented cartilage in clinical T2W sagittal knee images may not truly represent histologically defined deep and superficial zones. Furthermore, the normal to femoral cartilage surface varies considerably within imaging slices from the lateral to medial side of knee. Consequently, the standard magic angle effect (MAE) functions become inadequate for characterizing the observed anisotropic T2W signals in knee cartilage. Thus, a generalized MAE model is presented for better quantifying anisotropic T2W signals in the clinical studies of human knee articular cartilage.

2980		The morphology and T2* value changes of 21 knee joint cartilage sub-regions before and after marathon assessed by prototype software
		Ranxu Zhang¹, Xiaoshuai Chen¹, Jian Zhao¹, Ping Zhang¹, Xiaoyue Zhou², Baohai Yu¹, and Jianping Ding³
		¹Department of Radiology, The third Hospital of Hebei Medical University, Hebei Province Biomechanical Key Laboratory of Orthopedics, Shijiazhuang, Hebei 050051, China, Shijiazhuang City, China, ²Siemens Healthineers Ltd., Shanghai, 201318, China., Shanghai, China, ³Department of Radiology, Affiliated Hospital of Hangzhou Normal University,Hangzhou 310015,China, Hangzhou, China
		This study is designed to assess the changes of articular cartilage volume, thickness and T2* value before and after marathon in non-professional athletes by semi-automatic cartilage segmentation prototype software.All volunteers underwent MR scan before and after marathon and accepted follow-up observation.We confirmed that MR biochemical imaging combined with morphological changes can be used to assess the internal structure and general changes of cartilage. The anterior part of medial tibial plateau may be a high risk area for cartilage degeneration under the long-distance marathon.

2981		T1 and T1rho Relaxation in Equine Groove Model of Cartilage Damage
		Olli Nykänen¹, Nina E Hänninen^1,2, Swetha Pala¹, Ali Mohammadi¹, Mohammadhossein Ebrahimi^1,2, Nikae CR te Moller³, Harold Brommer³, Rene van Weeren³, Janne TA Mäkelä¹, Rami K Korhonen¹, Juha Töyräs^1,4,5, and Mikko J Nissi^1,2
		¹Department of Applied Physics, University of Eastern Finland, Kuopio, Finland, ²Research Unit of Medical Imaging, Physics and Technology, University of Oulu, Oulu, Finland, ³Department of Clinical Sciences, Faculty of Veterinary Medicine, Utrecht University, Utrecht, Netherlands, ⁴Diagnostic Imaging Center, Kuopio University Hospital, Kuopio, Finland, ⁵School of Information Technology and Electrical Engineering, The University of Queensland, Brisbane, Australia
		In this study, we examined the T1 and T1rho relaxation times of surgically damaged articular cartilage of equine carpal joints. Two different damage models (sharp and blunt grooves), inducing variation into the type of cartilage damage, were examined. The study revealed that adiabatic T1rho relaxation time is significantly increased in the bluntly damaged cartilage when compared to control cartilage. No significant differences between different damage models or between sharply damaged and control groups were seen for either T1 or T1rho relaxation. Moreover, both relaxation times correlated moderately (R≈0.65) with equilibrium modulus of articular cartilage.

2982		Clinical MSK Imaging and Challenges at 7T
		Garret M. Powell¹, Robert J. Spinner^2,3, Benjamin M. Howe¹, Matthew A. Frick¹, Andrew J. Fagan¹, Venkata V. Chebrolu⁴, Peter D. Kollasch⁴, Eric G. Stinson⁴, Joel P. Felmlee¹, and Kimberly K. Amrami^1,2
		¹Radiology, Mayo Clinic, Rochester, MN, United States, ²Neurologic Surgery, Mayo Clinic, Rochester, MN, United States, ³Orthopedic Surgery, Mayo Clinic, Rochester, MN, United States, ⁴Siemens Medical Solutions USA, Inc., Rochester, MN, United States
		Clinical MSK imaging at 7T is advantageous as a result of increased signal to noise ratio, contrast resolution, and spatial resolution compared to lower field strengths. Nevertheless, we have identified several challenges when performing MSK imaging at this field strength including inferior image uniformity, artifacts, coil limitations, and concealed pathology. The purpose of this exhibit is to introduce our early clinical experience with MSK imaging at 7T and illustrate the challenges we have identified.

2983		Quantitative correlation of BMD and T2 using high-resolution CT and MRI imaging of rabbit knee joints.
		Yang Xia¹, Farid Badar¹, and Sarah Salem¹
		¹Oakland University, Rochester, MI, United States
		Correlated application of high-resolution MRI T2 of intact rabbit knee joints with the aid of bone density properties measured by isotropic high-resolution µCT. A multi-model study will aid in the understanding of the relationship between underlying bone-density changes and osteoarthritic cartilage in joint degeneration

2984		HR-MAS 1H-NMR investigations of ovine Achilles tendon and rat rotator cuff tendon
		Anshuman Swain¹, Johannes Leisen², Muhammad A. R. Anjum¹, Joe Pearson³, Johnna S. Temenoff³, Felix M. Gonzalez¹, and David A. Reiter¹
		¹Department of Radiology and Imaging Sciences, Emory University, Atlanta, GA, United States, ²Department of Chemistry, Georgia Institute of Technology, Atlanta, GA, United States, ³Department of Biomedical Engineering, Georgia Institute of Technology, Atlanta, GA, United States
		Recent work quantifying tendon chemical shift information has reported resonances consistent with proteoglycan and lipid chemical groups. Currently, there is a lack of prior NMR studies of tendon available to interpret these signals and make assignments to matrix constituents. HR-MAS ¹H-NMR is an ideal tool for resolving chemical composition and structure in semisolid tissues like tendon. This study reports ¹H-NMR spectral assignments using 1D and 2D HR-MAS experiments in ex vivo tendon samples. Preliminary results show peaks attributed to residues comprising collagen fibrils, proteoglycans, lipids, and lactate. Ongoing work seeks to characterize ¹H NMR signatures of healthy and injured tendon.

2985		ATP Resynthesis after Exercise in Human Skeletal Muscle: Assessment by Phosphocreatine vs. Inorganic Phosphate Recovery in Elderly Subjects
		Jimin Ren^1,2, Craig R Malloy^1,2, Wanpen Vongpatanasin³, and Jarett Berry³
		¹Advanced Imaging Research Center, UT Southwestern Medical Center, Dallas, TX, United States, ²Department of Radiology, UT Southwestern Medical Center, Dallas, TX, United States, ³Department of Internal Medicine, UT Southwestern Medical Center, Dallas, TX, United States
		In this study, we investigated both phosphocreatine (PCr) and Pi as biomarkers of ATP resynthesis after exercise. We developed an efficient 1-min protocol of modest exercise, suitable for subjects in a wide range of age (from 20s to 80s), to generate sufficient large (~50%) PCr depletion and minimal pH drop (< 0.1) for quantitative measurement of ATP metabolism. We compared the rate constant for ATP production measured by Pi-based direct ATP synthase reaction (Pi + ADP → ATP) to PCr-based indirect CK-mediated reaction (ATP + Cr ↔ PCr + ADP), in the elderly population (aged 73±5 years).

Muscle MRI

Musculoskeletal

3165		Dynamic MR imaging of muscle contraction during electrical stimulation in facioscapulohumeral muscular dystrophy: a longitudinal study
		Xeni Deligianni^1,2, Francesco Santini^1,2, Matteo Paoletti³, Francesca Solazzo³, Niels Bergsland^4,5, Giovanni Savini³, Arianna Faggioli³, Giancarlo Germani³, Mauro Monforte⁶, Giorgio Tasca⁶, Enzo Ricci⁶, and Anna Pichiecchio^3,7
		¹Department of Radiology/ Division of Radiological Physics, University Hospital of Basel, Basel, Switzerland, ²Biomedical Engineering, University of Basel, Allschwil, Switzerland, ³Advanced Imaging and Radiomics Center, Neuroradiology Department, IRCCS Mondino Foundation, Pavia, Italy, ⁴Buffalo Neuroimaging Analysis Center, Department of Neurology, Buffalo Neuroimaging Analysis Center, Department of Neurology, Jacobs School of Medicine and Biomedical Sciences, Buffalo, NY, United States, ⁵IRCCS, Fondazione Don Carlo Gnocchi ONLUS, Milan, Italy, ⁶Unità Operativa Complessa di Neurologia, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy, ⁷Department of Brain and Behavioral Sciences, University of Pavia, Pavia, Italy
		Quantitative muscle MRI (T2&fat mapping) is progressively more used to assess disease involvement in muscle disorders, but in a static way, with dynamic assessment usually performed by clinical and instrumental examinations. To characterize muscle deformation we applied dynamic MRI synchronized with neuromuscular electrical stimulation evaluating the quadriceps muscles in 34 ambulatory subjects affected by facioscapulohumeral muscular dystrophy (FSHD) and ten healthy controls, longitudinally. Maximum values and rates of the strain differed between the two groups. Our results suggest that the evaluation of muscle ability to contract could be potentially used to monitor the evolution of muscle involvement in FSHD.

3166		Effect of Eddy Current Correction on Muscle Diffusion Measurements
		Xingyu Zhou^1,2, Melissa T. Hooijmans^1,3, Crystal L. Coolbaugh¹, Mark K. George¹, and Bruce M. Damon^1,2,4,5
		¹Vanderbilt University Institute of Imaging Science, Nashville, TN, United States, ²Department of Biomedical Engineering, Vanderbilt University, Nashville, TN, United States, ³Department of Radiology & Nuclear Medicine, Amsterdam Movement Sciences, Amsterdam University Medical Center, Amsterdam, Netherlands, ⁴Department of Radiology & Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, United States, ⁵Department of Molecular Physiology & Biophysics, Vanderbilt University Medical Center, Nashville, TN, United States
		Diffusion-tensor MRI (DTMRI) fiber tractography is a useful tool for reconstructing the structure of white matter tracts, nerves, and muscles. Eddy current correction is recognized as an important step in analysis of neuronal DTMRI data, while its effect on muscle DTMRI analysis has not been studied systematically. In this work we describe a preliminary comparison of muscle diffusion tensor imaging results with and without including eddy current correction in the pre-processing steps, including the morphological mismatch with anatomical image, the estimation of eigenvalues and fractional anisotropy, and the distribution of the first eigenvector within the selected muscle.

3167		Altered creatine kinase activity and mitochondrial oxidative capacity in muscular dystrophic mdx mice after repeated muscle contraction
		Kihwan Kim¹, Yuning Gu¹, Yuran Zhu¹, Yudu Li^2,3, Sherry Huang¹, Zhi-Pei Liang ^2,3, and Xin Yu^1,4
		¹Biomedical Engineering, Case Western Reserve University, cleveland, OH, United States, ²Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ³Beckman Institute for advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ⁴Case Center for Imaging Research, Case Western Reserve University, cleveland, OH, United States
		This study investigated the creatine kinase activity and the effects of repeated muscle contraction on muscle metabolism in mdx mice, a mouse model for Duchenne Muscular Dystrophy (DMD). Using phosphorous-31 (³¹P) magnetic resonance fingerprinting and dynamic ³¹P magnetic resonance spectroscopy, we observed a significant decrease in creatine kinase (CK) reaction rate in mdx mice compared to control mice. Decrease in phosphocreatine (PCr) recovery rate after its depletion by stimulation-induced muscle contraction was also observed in mdx mice. These findings suggest altered mitochondrial function in mdx mice.

3168		Establishing a computational approach to investigate the biophysical basis of relaxivity contrast imaging in the context of muscle degeneration
		Natenael B Semmineh¹, Sudarshan Ragunathan², Laura C Bell², and C Chad Quarles ²
		¹Imaging Research, Barrow Neurological Institute rrow Neurological Institute, Phoenix, AZ, United States, ²Imaging Research, Barrow Neurological Institute, Phoenix, AZ, United States
		The relaxivity contrast imaging parameter, TRATE (tissue transverse relaxivity at tracer equilibrium), has been shown to decrease during ALS induced myofiber degradation, due to T1 and T2* effects that occur when the contrast agent distributes around myofibers. The goal of this study is to expand a validated computational method to investigate the biophysical basis of TRATE in muscles of ALS patients and to optimize RCI acquisition parameters.

3169		Effect of moderate exercise on diffusion indices in skeletal muscle
		Amy R McDowell¹, Matthew T Lee², Kiran K Seunarine², and Chris A Clark²
		¹Centre for Neuromuscular Diseases, UCL Queen Square Institute of Neurology, London, United Kingdom, ²GOS UCL Institute of Child Health, London, United Kingdom
		An examination of the effect of moderate exercise on diffusion indices in skeletal muscle in volunteers in order to quantify the effect of the exercise, and define when diffusion indices returned to baseline rest values. We found that there was no significant difference between pre- and post-exercise MD or FA. We believe that moderate exercise will have no evident effect on diffusion indices, and no period of rest pre-scan is required.

3170		Quantification of gluteal and multifidus muscle asymmetry and fat infiltration in patients with unilateral lumbosacral nerve root compression
		Mengyue Wang¹, Weiqiang Dou², Yu Zheng¹, Yin Shi¹, and Yuefen Zou¹
		¹The First Affiliated Hospital of Nanjing Medical University, Nanjing, China, ²GE Healthcare, MR Research China, Beijing, China
		Muscle asymmetry and fatty infiltration resulting from lumbosarcal nerve root compression are associated with movement ability directly. However, the specific rule remains unclear. In this study, we used three dimensional fast imaging empolying steady-state acquisition with phase cycling(3D-FIESTA-C) and iterative decomposition of water and fat with echo asymmetry and least-squares estimation intelligent quantification (IDEAL-IQ) technique to evaluate the changes quantitatively. We found muscle changes were side- and level-specific, affected by several other factors. Muscle content is suggested as a good parameter for clinical evaluation.

3171		Interleaved multi-voxel 1H and 31P MRS for dynamic pH monitoring in the exercising calf muscle at 3 T with dynamic water reference.
		Alfredo Liubomir Lopez Kolkovsky^1,2, Beatrice Matot^1,2, Eric Giacomini¹, Martin Meyerspeer³, Benjamin Marty^1,2, and Harmen Reyngoudt^1,2
		¹NMR Laboratory, Institute of Myology, Paris, France, ²NMR Laboratory, CEA\DRF\IBFJ\MIRCen, Paris, France, ³High Field MR Center, CMPBME, Medical University of Vienna, Vienna, Austria
		We extended an interleaved ¹H/³¹P multi-voxel sequence for simultaneous pH monitoring based on carnosine (pH_1H, PRESS) and Pi (pH_31P, semi-LASER) with a co-localized dynamic water reference (PRESS). Simultaneous pH evaluations were done at 3 T in the soleus and gastrocnemius muscles during a straight-leg plantar flexion paradigm. A pH_1H bias of ~0.05 units was removed by using the dynamic water reference. pH_1H and pH_31P time courses (1-minute temporal resolution) were comparable. Pi and Carnosine (C2-H) resonances broadened during exercise in the gastrocnemius. This method opens the way for new applications in functional studies targeting individual muscles at 3 T.

3172		A Marker Controlled Active Contour Model for Thigh Muscle Segmentation in MR Images
		Weihong Guo¹, Michael Judkovich¹, Richard Lartey², Dongxing Xie², Mingrui Yang², and Xiaojuan Li²
		¹Mathematics, Applied Mathematics and Statistics, Case Western Reserve University, Cleveland, OH, United States, ²Biomedical Engineering, Program of Advanced Musculoskeletal Imaging (PAMI), Lerner Research Institute, Cleveland Clinic, Cleveland, OH, United States
		Thigh muscle morphology and composition quantified from MR images are potential imaging biomarkers for diseases such as osteoarthritis and sarcopenia. MR thigh muscle segmentation is an important step in quantifying both muscle morphology and composition. Unfortunately, the thigh muscle groups are tightly bundled together, making them very hard to segment due to a lack of clear boundaries between different muscles. We proposed a novel geometric flow based semi-automatic scheme to effectively segment them. We combined reproducible kernel Hilbert space edge descriptor and geodesic distance maps from a set of markers and anti-markers to define the force for the geometric flow.

3173		DT-MRI with the random permeable barrier model shows larger, more heterogeneous muscle fibre diameters in Becker muscular dystrophy patients
		Donnie Cameron^1,2, Jedrzej Burakiewicz¹, Nienke M. van de Velde³, Celine Baligand¹, Thom T.J. Veeger¹, Melissa T. Hooijmans⁴, Jan J.G.M. Verschuuren³, Erik H. Niks³, and Hermien Kan¹
		¹C.J. Gorter Centre for High Field MRI, Department of Radiology, Leiden University Medical Centre, Leiden, Netherlands, ²Norwich Medical School, University of East Anglia, Norwich, United Kingdom, ³Department of Neurology, Leiden University Medical Centre, Leiden, Netherlands, ⁴Biomedical Engineering and Physics, Amsterdam University Medical Centre, Amsterdam, Netherlands
		Becker muscular dystrophy (BMD) is characterised by progressive muscle damage and weakness. Diffusion tensor imaging (DTI) represents a promising candidate for studying BMD; the random permeable barrier model (RPBM), in particular, gives estimates of muscle fibre diameters and membrane permeabilities. Here we study DTI and RPBM metrics in BMD patients and controls. Spin-echo and stimulated-echo DTI data were acquired in the lower leg and RPBM metrics calculated. DTI metrics showed time-dependence and intramuscular differences, but no inter-group differences. RPBM analysis, however, successfully showed differences between BMD patients and controls, with fibre diameters being larger and more variable in patients.

3174		Optimized water T2 mapping from multi-echo spin-echo acquisitions with an external fat fraction constraint
		Francesco Santini^1,2, Xeni Deligianni^1,2, Matteo Paoletti³, Francesca Solazzo³, Matthias Weigel^1,4,5, Paulo Loureiro De Sousa⁶, Oliver Bieri^1,2, Mauro Monforte⁷, Enzo Ricci^7,8, Giorgio Tasca⁷, Anna Pichiecchio^3,9, and Niels Bergsland^10,11
		¹Radiological Physics, Department of Radiology, University Hospital of Basel, Basel, Switzerland, ²Department of Biomedical Engineering, University of Basel, Allschwil, Switzerland, ³Neurological Institute Foundation Casimiro Mondino (IRCCS), Pavia, Italy, ⁴Translational Imaging in Neurology, Department of Biomedical Engineering, University of Basel, Allschwil, Switzerland, ⁵Department of Neurology, University Hospital of Basel, Basel, Switzerland, ⁶UMR7357 Laboratoire des sciences de l'Ingénieur, de l'Informatique et de l'Imagerie (ICube), Strasbourg, France, ⁷Institute of Neurology, A. Gemelli University Hospital Foundation, Catholic University of the Sacred Heart, Rome, Italy, ⁸Institute of Neurology, Catholic University of the Sacred Heart, Rome, Italy, ⁹Department of Brain and Behavioral Sciences, University of Pavia, Pavia, Italy, ¹⁰Western New York Stem Cell Culture and Analysis Center, Jacobs School of Medicine and Biomedical Sciences, University of Buffalo, Buffalo, NY, United States, ¹¹Fondazione Don Carlo Gnocchi Onlus (IRCCS), Milan, Italy
		The measurement of the T2 of the water component is an important marker of neuromuscular diseases. Strategies exist to estimate it from conventional multi-echo spin-echo acquisitions, taking into account fat infiltration and B1 inhomogeneity. In this work, we present an open software that is able to efficiently perform this kind of fitting, and additionally to incorporate the information from an external fat fraction acquisition, for increased accuracy and reduced scan time.

3175		Elliptical filter optimization for HARP based strain quantification in skeletal muscle
		Melissa T. Hooijmans^1,2, Crystal L. Coolbaugh³, Xingyu Zhou^2,4, Mark K. George², and Bruce M. Damon^4,5,6
		¹Department of Radiology & Nuclear Medicine, Amsterdam Movement Sciences, Amsterdam UMC, Location AMC, Amsterdam, Netherlands, ²Vanderbilt University Institute of Imaging Science, Nashville, TN, United States, ³Vanderbilt University Institute of Imaging Sciences, Nashville, TN, United States, ⁴Department of Biomedical Engineering, Vanderbilt University Medical Center, Nashville, TN, United States, ⁵Department of Radiology & Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, United States, ⁶Department of Molecular Physiology & Biophysics, Vanderbilt University Medical Center, Nashville, TN, United States
		SPAMM, combined with HARP, is primarily used for strain quantification in the myocardium; but applications in other soft tissues, including skeletal muscle, are increasing. HARP suffers from artifacts due to usage of inappropriate filters and should therefore be optimized especially for skeletal muscle. We simulated strains ranging from -0.156 to +0.156 (decimal strain) in dynamically acquired line- and grid-tagged SPAMM images, and optimized elliptical filter parameters were determined for skeletal muscle. With this filter, differences between the measured strain and absolute strain were small for the low strain values and increased with the actual strain values and number of dynamics.

3176		Imaging of calf muscle contraction in pediatric patients with cerebral palsy: comparison of voluntary motion and electrically evoked motion
		Claudia Weidensteiner^1,2, Xeni Deligianni^1,2, Tanja Haas¹, Philipp Madoerin¹, Oliver Bieri^1,2, Meritxell Garcia³, Jacqueline Romkes⁴, Erich Rutz⁵, Francesco Santini^1,2, and Reinald Brunner⁶
		¹Department of Radiology, Division of Radiological Physics, University Hospital Basel, Basel, Switzerland, ²Department of Biomedical Engineering, University of Basel, Basel, Switzerland, ³Department of Radiology, Division of Neuroradiology, University Hospital Basel, Basel, Switzerland, ⁴Laboratory for Movement Analysis, University Children's Hospital Basel, Basel, Switzerland, ⁵Hugh Williamson Gait Laboratory, The Royal Children's Hospital Melbourne, Parkville, Australia, ⁶Department of Orthopedic Surgery, University Children's Hospital Basel, Basel, Switzerland
		Aim of this study is to investigate the feasibility of phase contrast imaging at 3T for assessment of muscle function in children with cerebral palsy. Time-resolved cine phase contrast MRI was synchronized with (a) electrical muscle stimulation (EMS) of the calf muscle (b) voluntary plantarflexion following visual instructions. The achieved force was higher for the voluntary task, but its periodicity was lower compared to the stimulated contraction. Therefore, it was not possible to resolve two distinct velocity peaks for voluntary contraction and release, but it was during EMS, in case the tolerated current was high enough to evoke sufficient contraction.

3177		IDEAL-CPMG muscle T2-water and Dixon Fat-Fraction Maps in People with Inclusion Body Myositis and Charcot-Marie-Tooth disease Type 1A
		Amy R. McDowell¹, Stephen J. Wastling^1,2, Lara Cristiano³, Jasper M. Morrow¹, Matthew R.B. Evans⁴, Christopher D.J. Sinclair¹, Pedro M. Machado¹, Michael Hanna¹, Mary M. Reilly¹, Tarek Yousry¹, and John S. Thornton¹
		¹MRC Centre for Neuromuscular Diseases, UCL Queen Square Institute of Neurology, London, United Kingdom, ²Lysholm Department of Neuroradiology, National Hospital for Neurology and Neurosurgery, London, United Kingdom, ³Department of Pediatric Neurology and Radiology, Fondazione Policlinico Universitario, Rome, Italy, ⁴St Thomas Hospital, London, United Kingdom
		Muscle T₂ and fat-fraction are two promising metrics for sensitive measures of early disease-related changes in muscle for use in clinical trials. However, fat has a longer T₂ than water in muscle and masks underlying increases in water T₂ due to muscle oedema. The IDEAL-CPMG sequence with appropriate image-data processing produces a T₂-water map uncontaminated by the fat signal. This sequence was acquired in healthy volunteers and two different muscular dystrophy disease types. It was found that elevated T₂-water may be a predictor of later progression to fatty-atrophy in muscle, supporting the value of future longitudinal studies to test this.

3178		Denoising dynamic CrCEST in skeletal muscle following exercise using low rank tensor approximations
		Neil Wilson¹, Mark A Elliott², Dushyant Kumar², and Ravinder Reddy²
		¹Siemens Medical Solutions USA Inc, Malvern, PA, United States, ²CMROI, University of Pennsylvania, Philadelphia, PA, United States
		Dynamic CrCEST allows for high resolution mapping of the creatine kinase reaction recovery kinetics following exercise. However, voxel averaging over large ROIs is usually done to get reliable fits because of high variation due to limited SNR. Here, we show that improved denoising utilizing low rank tensor approximations that exploit the full dimensionality of the scans allows for reliable fits over smaller volumes or even single voxels, making muscle response heterogeneity measurable.

3179		Feasibility Study of 31P Multivoxel Spectroscopy for the detection of Alkaline Inorganic Phosphate in multiple compartments of the lower leg at 3T
		Rajakumar Nagarajan¹ and Gwenael Layec²
		¹Human Magnetic Resonance Center, Institute for Applied Life Sciences, University of Massachusetts Amherst, Amherst, MA, United States, ²Kinesiology, University of Massachusetts Amherst, Amherst, MA, United States
		In this study, we demonstrate the feasibility and sensitivity of ³¹P two-dimensional CSI sequence for the detection of alkaline Pi in the soleus, tibialis anterior, and gastrocnemius muscles in six healthy volunteers in rest. The major finding of the current study was the consistent detection of an alkaline Pi from the cytosolic Pi signal in the volume localized ³¹P MRS from the resting muscle in healthy human subjects at 3T using decoupling technique. This study provides the proof of concept for a non-invasive and localized method to determine the alkaline Pi, a potential index of mitochondrial density at rest.

3180		Measuring motor unit morphology in upper extremities using motor unit magnetic resonance imaging (MUMRI)
		Matthew Birkbeck^1,2,3, Linda Heskamp¹, Ian Schofield¹, Roger Whittaker¹, and Andrew Blamire¹
		¹Translational and Clinical Research Unit, Newcastle University, Newcastle upon Tyne, United Kingdom, ²Newcastle Biomedical Research Centre, Newcastle Biomedical Research Centre, Newcastle upon Tyne, United Kingdom, ³Northern Medical Physics and Clinical Engineering, Newcastle upon Tyne NHS Foundation Trust, Newcastle upon Tyne, United Kingdom
		Motor unit (MU) magnetic resonance imaging (MUMRI) is a non-invasive technique which detects muscle fibre micro-contraction and is based on diffusion weighted MRI. To date MUMRI has been applied in conjunction with in scanner electrical stimulation to study MU activity in the lower leg. Here we present the first use of MUMRI in the upper limbs to study single human MUs. The acquired images show low levels of distortion and good fat suppression, allowing single human MU sizes and shapes to be determined. This is of interest in neuromuscular diseases as it is a non-invasive way to study MU morphology.

3181		MR analysis of thigh muscle myopathy using texture features and supervised machine learning
		Hon J Yu¹, Saya Horiuchi^1,2, Toshimi Tando¹, Vincent J Caiozzo³, Virginia E Kimonis⁴, and Hiroshi Yoshioka¹
		¹Radiological Sciences, University of California, Irvine, Orange, CA, United States, ²Radiology, St. Luke's International Hospital, Tokyo, Japan, ³Department of Orthopaedics, Physiology & Biophysics, University of California, Irvine, Irvine, CA, United States, ⁴Division of Genetic and Genomic Medicine, Department of Pediatrics, University of California, Irvine, Irvine, CA, United States
		This study evaluates texture features to demonstrate their relationship with muscle classification based on a 5-grade scale and their value as classifier when trained in supervised machine-learning framework. The results suggest the texture features capture various image characteristics that are likely utilized during manual muscle classification by human being and can correctly predict with up to 81% accuracy when properly trained in supervised machine-learning setting. A further study with bigger data size would be necessary to fully examine such classification model and also to look into the possibility of selecting subset of features to make such an approach more practical.

3182		1H MRS: a tool to study age-related changes in intracellular metabolites in the deep spinal muscles?
		Sarah Catherine Wayte¹, Alexander Dallaway², Andrew David Weedall¹, John Hattersley³, and Adrian John Wilson^3,4
		¹Radiology Physics, Department of Clinical Physics and Bioengineering, University Hospital, Coventry, United Kingdom, ²Centre for Sport, Exercise and Life Sciences, Coventry University, Coventry, United Kingdom, ³Coventry NIHR CRF, Human Metabolic Research Unit, University Hospital, Coventry, United Kingdom, ⁴Department of Physics, University of Warwick, Coventry, United Kingdom
		¹H MRS was used to investigate relative concentrations of methylene intracellular lipids [IMCL_CH2]_R and creatine [Cr]_R in the psoas of groups of younger (19-30yrs) and older (61-81yrs) fit male volunteers. No difference was found between groups for either metabolite (p>0.05). However, the median value for [Cr]_R in the older group (median;Q1-Q3=1.02;0.55-1.29%) was much higher than in the younger group (0.57;0.50-0.60%) suggesting [Cr]_R could partially explain reduction in muscle strength with age being greater than predicted by loss of muscle mass. There was no correlation between metabolite concentrations (r²=0.21) so [Cr]_R may possibly be a marker for age-related changes in muscle.

3183		SEMI-AUTOMATIC QUANTIFICATION OF ABDOMINAL WALL MUSCLES DEFORMATIONS BASED ON DYNAMIC MRI IMAGE REGISTRATION
		Arthur Jourdan¹, Arnaud Le Troter², Pierre Daude², Stanislas Rapacchi², Catherine Masson¹, Thierry Bege^1,3, and David Bendahan²
		¹Aix-Marseille Univ, Univ Gustave Eiffel, IFSTTAR, LBA, Marseille, France, ²Aix Marseille Univ, CNRS, CRMBM, Marseille, France, ³Department of General Surgery, Aix Marseille Univ, North Hospital, APHM, Marseille, France
		A novel semi-automatic post-processing method dedicated to real-time dynamic MRI aiming at a fast and reliable quantification of abdominal wall muscles deformations. The described method combines a highly accurate supervised 2D+t segmentation procedure of the abdominal wall muscles (mean Dice similarity coefficient of 0.95 ± 0.03), the quantification of muscles deformations based on masks registration and the mapping of deformations within muscle compartments leveraging a dedicated parcellation. The present genuine method provides a quantitative analytical frame that could be used in further studies for a better understanding of abdominal wall deformations in physiological and pathological situations.

3184		Feasibility and Reproducibility Study of Diffusion-Tensor Imaging in Rotator Cuff Muscles of Asymptomatic Volunteers.
		Cyril Tous, PhD¹, Alexandre Jodoin, MD², Detlev Grabs, MD, PhD³, Elijah Van Houten, PhD⁴, and Nathalie J Bureau, MD MSc FRCP(C)^1,2
		¹Radiology, Centre de recherche du Centre hospitalier de l’Université de Montréal, Montreal, QC, Canada, ²Radiology, Centre hospitalier de l’Université de Montréal, Montréal, QC, Canada, ³Anatomy, Université du Québec à Trois-Rivières, Trois-Rivières, QC, Canada, ⁴Mechanical Engineering, Université de Sherbrooke, Sherbrooke, QC, Canada
		Surgical planning of rotator cuff tears does not benefit from quantitative measurement of muscles stiffness and microstructure leading to retears in the following years. Since stiffness is dependent on anisotropy, Diffusion Tensor Imaging can simplify the inversion algorithm of Magnetic resonance elastography. Repeatability of DTI metrics from three scans in six shoulders of six asymptomatic volunteers was achieved (coefficient of variance <10%) for the fractional anisotropy, mean diffusivity, mode and eigenvalues. Clear convergence of myocytes to the tendons was observed with tractography in the supraspinatus as known in dissection. This MRI protocol is promising for developing biomarkers for surgical planning.

Spine

Muscle MRI

Musculoskeletal

3185		Clinical feasibility of ultrafast lumbar spine magnetic resonance imaging: a preliminary report
		Nobuo Kashiwagi¹, Yuichi Yamashita², Hitoshi Watanabe³, Hisashi Tanaka¹, Katsusuke Kyotani², Hiroto Takahashi¹, Chisato Matsuo¹, Takehisa Sakisuka¹, Azusa Miura¹, Masahiro Fujiwara¹, Atsuko Arisawa ¹, and Noriyuki Tomiyama¹
		¹Osaka University Graduate School of Mediine, Suita, Japan, ²Canon Medical Systems Corporation, Otawara, Japan, ³Department of Radiology, Yukoukai General Hospital, Ibaraki, Japan
		Using deep learning reconstruction, we developed an ultrafast lumbar MRI protocol with a total acquisition time of 1 minute 53 seconds. Compared with the standard MRI protocol, the interprotocol agreement for the classification of degenerative lumbar changes was almost perfect or substantial agreement with weighted kappa values of 0.72 to 0.84. The two protocols showed 100% concordance for the detection of clinically significant pathologies. Our preliminary study demonstrated that the ultrafast lumbar MRI protocol adequately preserved diagnostic equivalency with the standard protocol, suggesting that it could be an alternative protocol for restless patients and in an emergency setting.

3186		High-resolution head and C-spine MRI using a Zero-TE (ZTE) sequence at 3T
		Aiming Lu¹, Carrie M Carr¹, Norbert G Campeau¹, Steven A Messina¹, and David F Kallmes¹
		¹Radiology, Mayo Clinic, Rochester, MN, United States
		Zero TE (ZTE) sequences have been demonstrated to be capable of achieving CT-like images of the skull and C-spine. To obtain high-resolution images of the skull and c-spine, two separate acquisitions are often performed. Although in general cortical bone, soft tissue and air could be differentiated based on the signal intensity, fatty tissues and connective tissues can be problematic due to the chemical shift artifacts and/or having similar signals to that of cortical bones. This work demonstrated that with optimized protocol and advanced post-processing, ZTE MRI allows for high quality skull and C-spine images in a single acquisition.

3187		A Novel Hybrid Score Integrating Clinical and MR Features to Predict Lumbar Vertebral Osteoporosis in Female
		Shao-Chieh Lin¹, Chun-Jung Juan², Yi-Jui Liu³, Chun-Wen Chen⁴, Chien-Yuan Wang⁵, Wu-Chung Shen⁶, Der-Yang Cho⁷, and Kai-Yuan Cheng⁸
		¹Ph.D. program in Electrical and Communication Engineering, Feng Chia University, Taichung, Taiwan, ²Department of Medical Imaging, China Medical University Hsinchu Hospital, Hsinchu, Taiwan, ³Department of Automatic Control Engineering, Feng Chia University, Taichung, Taiwan, ⁴Department of Radiology, Taichung Armed Forces General Hospital, Taichung, Taiwan, ⁵Department of Orthopedics, China Medical University Hsinchu Hospital, Hsinchu, Taiwan, ⁶Department of Medical Imaging, China Medical University Hospital, Taichung, Taiwan, ⁷Department of Neurosurgery, China Medical University Hospital, Taichung, Taiwan, ⁸Department of Medical Imaging and Radiological Sciences, Central Taiwan University of Science and Technology, Taichung, Taiwan
		Lumbar vertebral osteoporosis (LvOPO) in women is clinically important. It is affected by mixed causes, including but not limited to the aging, lumbar vertebral fat fraction (LvFF), and estrogen. In the current study, the LvOPO in women was diagnosed by a newly proposed hybrid scoring system, i.e., the LvOPO score (LvOPOS). By integrating the independent predictors including age, LvFF, and menopause-MR interval (MMI), the LvOPOS achieved an AUC higher than the LvFF (0.802), age (0.794), and MMI (0.788). Our results provide a new insight in diagnosing the LvOPO specific to women with relevant to both clinical and MR features.

3188		A comparative study of R2, UTE-T2 and T1rho mapping for evaluation of degenerative alterations in human intervertebral discs
		LiLan Wu¹, JianJun Zhou², and Pu-Yeh Wu³
		¹Department of Radiology, Fudan University Affiliated Zhongshan Hospital Xiamen Branch, Xiamen, China, ²Department of Radiology, Fudan University Affiliated Zhongshan Hospital, Shanghai, China, ³GE Healthcare, Beijing, China
		This study aims to investigate whether human lumbar IDD can be detected using R2, UTE-T2 and T1rho mapping. We systematically compared the efficacy of R2, UTE-T2 and T1rho values in the diagnosis of early IDD. Specifically, we found that T1rho value is superior to UTE-T2* and R2* values for diagnosis of the early IVD, while UTE-T2* value is optimal for diagnosis of advanced IVD. Overall, we concluded that R2* and UTE-T2* mapping provides another promising method for quantitatively evaluated lumbar IDD, and T1rho mapping can be considered an effective tool for distinguishing IDD at earlier stage of the degenerative process

3189		Normal fetal development of the cervical, thoracic and lumbar spine: a post-mortem study based on magnetic resonance imaging
		Shuai Zhang¹, Xiangtao Lin², Ximing Wang², Xiang Feng³, and Rui Diao²
		¹School of Medicine, Shandong First Medical University, School of Medicine, Shandong First Medical University, Jinan, China, ²Department of Radiology, Shandong Provincial Hospital affiliated to Shandong First Medical University, Jinan, China, ³MR Scientific Marketing, Diagnosis Imaging, Siemens Healthcare Ltd, Beijing, China
		Before evaluating spinal pathology, it is essential to have knowledge of the normal spinal development at different gestational ages. Postmortem magnetic resonance imaging was performed on 55 fetuses (gestational ages, 17–42 weeks) by using three-dimensional T2-weighted sequences. The volumes of interest were manually outlined for the cervical, thoracic, lumbar, and L1–L5 centrum ossification centers (COCs), and the COC volumes (COCVs) were calculated. The cervical, thoracic, and lumbar COCVs showed a positive relationship with gestational age.The cervical, thoracic, lumbar, and L1–L5 COCVs show good correlation with gestational age in the second and third trimesters.

3190		Spinal Giant Cell Tumor of Bone: Immunohistochemistry and Preoperative Magnetic Resonance Imaging Features for Prognostic Prediction
		Qizheng Wang¹, Siyuan Qin¹, Yang Zhang², Enlong Zhang³, Xiaoying Xing¹, Min-Ying Su², and Ning Lang¹
		¹Radiology, Peking University Third Hospital, Beijing, China, ²Center for Functional Onco-Imaging, Irvine, CA, United States, ³Radiology, Peking University International Hospital, Beijing, China
		A new evaluating system based on tumor immunohistochemistry and preoperative MRI features in spinal giant cell tumor of bone to predicting overall survival of total en bloc spondylectomy patients with over 2 years follow up. The largest lesion diameter (>4.2 cm) and the vertebral compression were independent predictors of postoperative recurrence. According to Kaplan-Meier survival analysis, the cystic change in the lesion and the degree of compression ≥50% suggest a worse clinical outcome. The expression levels of vascular endothelial growth factor and p53 gene have no obvious clinical significance on the survival outcome. H3F3A was positively expressed in our cohort.

3191		Effects of different acceleration factors on lumbar body fat quantification using 3D mDixon Quant technique
		Yu Song¹, Qingwei qing Song², Yingkun Guo¹, Gang Ning¹, and Xuesheng Li¹
		¹Department of Radiology, West China Second University Hospital, Sichuan University, Chengdu, China, ²Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian, China
		Compressed SENSE (CS) is a cutting-edge signal acquisition and processing innovation technology based on applied mathematics. By directly collecting the compressed signal, the amount of sampled data is greatly reduced, and the optimal acceleration factor can be passed. The realization of the method can shorten the clinical magnetic resonance scanning time while ensuring the spatial resolution.

3192		Quantitative T2 parametric value in assessment of abnormal lumbar paraspinal muscle in patients with low back pain
		Yinqi Liu¹, Huiting Deng¹, Weiyin Liu², and Kun Zhang^1,3
		¹Department of Radiology, First Affiliated Hospital of Hunan University of Chinese Medicine, Changsha, China, ²MR Research, GE Healthcare, Beijing, China, ³College of Integrated Traditional Chinese and Western Medicine, Hunan University of Chinese Medicine, Changsha, China
		Chronic low back pain (CLBP) has been reported to be associated with low back dysfunction with the characteristics of decreased cross-section area (CSA) of paraspinal muscles and increased adipose content. In our study, T2 value at L4/L5 multifidus muscle was significantly higher in LBP group than in HCs. The multifidus muscle is the most medial in the low back extensor group and has the maximum strain stress when spinal movement and balance maintenance, so it might easily cause the abnormality of the low back muscle group. T2 mapping could be a good predictor of detecting minimal variation of multifidus muscles.

3193		Evaluation of Lumbar Disc Degeneration Using Ultrashort Echo Time Magnetization Transfer (UTE-MT) Imaging in Cartilaginous Endplate
		Jin Liu¹, Yajun Ma², Jianwei Liao¹, Xiaojun Chen¹, Wei Li¹, Lin Yao¹, Zhihai Su³, Long Qian⁴, Jiang Du², and Shaolin Li¹
		¹Department of Radiology, The Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, China, ²Department of Radiology, University of California, San Diego, CA, United States, ³Department of Orthopedics, The Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, China, ⁴MR Research, GE Healthcare, Guangzhou, China
		Ultrashort echo time magnetization transfer (UTE-MT) sequence can quantify magnetization transfer ratio (MTR) values of short T2 tissues, which has potential to evaluate early disc degeneration. The current study aims to assess the feasibility of UTE-MT quantification of cartilaginous endplate for evaluation of early disc degeneration. In addition, the latent associations between the UTE-MTR and grade of disc degeneration, age, gender, the oswestry disability index (ODI) were evaluated, respectively. Our results demonstrated that the UTE-MT sequence is feasible to quantify cartilaginous endplate in lumbar spine. UTE-MTR may be a potentially useful biomarker in the assessment of early lumbar disc degeneration.

3194		BOLD and IDEAL-IQ evaluation for lumbarparaspinal muscle changes in patients with chronic low back pain
		Zeng Xiao min¹, Huang Yi long², Nie Li sha³, and He Bo¹
		¹The First Affiliated Hospital of Kunming Medical University, Kunming, China, ²The First Affiliated Hospital of Kunming Medical University, kunming, China, ³GE healthcare, China, Beijing, China
		This study analyzed the relationship between the cross-sectional area (CSA), effective lateral relaxation rate (R2* value), fat fraction (FF value) of lumbar paraspinal muscles of patients with chronic low back pain (CLBP group) and asymptomatic control group (Control group). The multifidus muscle and erector spinae muscle at the central level of the L4/L5 intervertebral disc were the target muscles.BOLD and IDEAL-IQ provided a new idea for imaging evaluation of lumbar paraspinal muscles. Our study concluded that BOLD and IDEAL-IQ are feasible to evaluate the changes of lumbar muscle morphology and function in patients with CLBP.

3195		High-Contrast Lumbar Spinal Bone Imaging Using a 3D Slab-Selective UTE Sequence
		Amir Masoud Afsahi¹, Zhao Wei¹, Michael Carl², Saeed Jerban¹, Hyungseok Jang¹, Nicole Le¹, Jiang Du¹, Eric Y. Chang^1,3, and Ya-Jun Ma¹
		¹Department of Radiology, University of California, San Diego, San Diego, CA, United States, ²GE HealthCare, San Diego, CA, United States, ³Radiology Service, Veterans Affairs, San Diego Healthcare System, San Diego, CA, United States
		To explore the application of a 3D slab-selective UTE sequence as a superior technique to 3D ZTE at 3T field strength and to compare with 3D CT as a gold standard in Lumbar spinal bone assessment.

3196		Multimodal MRI evaluation of paraspinal muscles and TNF- α in rats with discogenic low back pain after rehabilitation training
		Kaiwen Yang¹, Baofa Luo¹, Yilong Huang¹, Lisha Nie², and Bo He¹
		¹Medical imaging department, The First Affiliated Hospital of Kunming Medical University, Kunming, China, ²GE Healthcare,MR Research China, Beijing, China
		Based on functional MRI, the current study aims to explore the pathological morphology, fat infiltration and the expression of TNF-α in paraspinal muscles of discogenic low back pain(DLBP）. And to further explore the relation of rehabilitation exercise on paraspinal muscles and intervertebral discs in rats. It is concluded that rehabilitation exercise can reduce the content of TNF-α in serum and paraspinal muscles, relieve the symptoms of DLBP in rats, and improve paraspinal muscle dysfunction.

3197		DCE-MRI for the Assessment of Outcome of CyberKnife Stereotactic Radiosurgery for Patients with Spinal Metastases
		Yongye Chen¹, Enlong Zhang ², Qizheng Wang ¹, Huishu Yuan¹, Huishu Yuan¹, Hongqing Zhuang¹, and Ning Lang¹
		¹Peking University Third Hospital, Beijing, China, ²Peking University International Hospital, Beijing, China
		The imaging methods used to evaluate the efficacy of radiosurgery in spinal metastases have certain limitations. DCE-MRI can help determine the vascularity and hemodynamics of tumors in vivo, and can be used to evaluate local tumor response. Twenty-seven patients with 39 lesions were included. Post-treatment Kep, ΔKtrans and ΔKep in the non-progressive disease group were significantly lower than the corresponding values in progressive disease group. Post-treatment Ve and ΔVe in the non-progressive disease group were significantly higher than that of the progressive disease group. ΔKtrans had the highest diagnostic efficiency, with an AUC of 0.821.

3198		Comparison of T2 relaxation time from synthetic MRI and conventional T2 mapping in the diagnosis of intervertebral disc degeneration
		Xiaoqing Liang¹, Weiyin Vivian Liu², Yitong Li¹, and Xiaoming Li³
		¹Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, ²MR Research, GE Healthcare, Beijing, China, ³1Radiology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
		Synthetic MRI has become a recent research hotspot due to its advantages of short-time acquiring multiple sets of weighted images and quantitative maps in one time. Synthetic MRI has been successfully used for the diagnosis of many cerebral diseases, but its feasibility and stability in other tissues and lesions need to be further studied. The current study demonstrates that the utility of synthetic MRI is slightly better than conventional T2 mapping in diagnosing intervertebral disc degeneration. In addition, one-time MAGiC scan on lumbar vertebrae should be concerned for clinical diagnosis.

3199		Histogram analysis of IDEAL-IQ and UTE images better assists to detect abnormalities of the lumbar vertebral bone marrow in patients with different stages of chronic kidney disease
		Yan Xiong¹, Weiyin Vivian Liu ², Fan He¹, Yanan Wang¹, Yao Zhang¹, Shuang Hu¹, and Xiaoming Li¹
		¹Wuhan Tongji Hospital, Wuhan, China, ²MR Research, GE Healthcare, Beijing, China
		Mineral bone disease (MBD) has an extremely high incidence in patients with chronic kidney disease (CKD) and significantly affects the prognosis and quality of life of patients. However, the current diagnostic methods are difficult to determine the type and severity of CKD-MBD. In our study, histogram analysis of IDEAL-IQ showed kurtosis and R2* could distinguish bone marrow changes in patients with CKD at different stages in indication of altered skeletal health and bi-component analysis of UTE images demonstrated short T2* value possibly in reflection of trabecular bone was highly correlated with blood biochemical indicators.

3200		Automated differentiation between benign and malignant vertebral compression fracture using a deep convolutional neural network on MRI
		Takafumi Yoda¹, Satoshi Maki², Koji Matsumoto¹, Hajime Yokota³, Yoshitada Masuda¹, and Takashi Uno³
		¹Department of Radiology, Chiba University Hospital, Chiba, Japan, ²Department of Orthopedic Surgery, Graduate School of Medicine, Chiba University, Chiba, Japan, ³Diagnostic Radiology and Radiation Oncology, Graduate School of Medicine, Chiba University, Chiba, Japan
		The Differentiating between osteoporotic vertebral fractures (OVFs) and malignant vertebral compression fractures (MVFs) due to spinal metastasis is a challenging problem for the spine surgeons and the radiologists. We evaluated the performance of our CNN model in differentiating between OVFs and MVFs on short-TI inversion recovery (STIR) and T1-weighted (T1WI) images compared with the performance of three spine surgeons. The sensitivity, specificity, and accuracy of the CNN on both STIR and T1WI were equal to or better than those of the three spine surgeons.

3201		The assessment of cartilage endplate and its relationship with the corresponding disc degeneration using UTE imaging
		Zhilin Ji¹, Weiqiang Dou², Yuefen Zou¹, Yin Shi¹, Yu Zheng¹, and Hongyuan Ding¹
		¹The First Affiliated Hospital of Nanjing Medical University, Nanjing, China, ²GE Healthcare, MR Research China, Beijing, P.R. China, Beijing, China
		In this study, we aimed to investigate the feasibility of ultra-short echo time (UTE) imaging for assessing the cartilage endplate (CEP) damage and evaluate the relationship between the grading of CEP and lumbar disc degeneration. UTE images with short and long TEs were performed on patients with low back pain. The resultant subtracted UTE images between short and long TEs showed well structural CEP clearly. Strong correlation was observed between CEP degree based on UTE imaging and lumbar intervertebral disc degeneration. With these findings, UTE imaging might thus be considered an effective tool to assess the CEP damage in clinic.

3202		Evaluation of lumbar facet joint and intervertebral disc degeneration using histogram analysis of T2 and T2＊values
		Xiaoqing Liang¹, Weiyin Vivian Liu², and Xiaoming Li¹
		¹Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, ²MR Research,GE Healthcare, Beijing, China
		Lumbar intervertebral disc (IVD) and lumbar facet joint (LFJ) are the primary and important load-bearing structures in spine. Any abnormality of the anterior IVD and the two posterior LFJs may lead to spinal instability and low back pain (LBP). The current study demonstrates that LFJ degeneration occurs earlier and progresses more slowly than IVD degeneration. Histogram analysis of T2 and T2* values is feasible for detecting and grading IVD but not LFJ degeneration.

3203		Fetal magnetic resonance imaging of normal lumbar spine development in vivo
		Xing Yin¹, Xin Zhao¹, Liying Zhang¹, Qingna Xing¹, Rui Yuan², and Zhijun Niu³
		¹Radiology, the Third Affiliated Hospital of Zhengzhou University, Zhengzhou, China, ²Ultrasound, the Third Affiliated Hospital of Zhengzhou University, Zhengzhou, China, ³Obstetrics and Gynecology, the Third Affiliated Hospital of Zhengzhou University, Zhengzhou, China
		In the last few years, fetal magnetic resonance (MR) has been more and more used in the obstetrics and gynecology. We measure the height and length of each lumbar vertebra body ossification center and the height of each intervertebral gap, and our results show a good linear correlation between normal lumbar vertebra body ossification center development and gestational age in vivo

3204		Diffusion Kurtosis Imaging and Intravoxel Incoherent Motion in diagnosis of active sacroiliitis with ankylosing spondylitis
		Xiaolin Guo^1,2, lixia Qian², Yueluan Jiang³, and Zhongshuai Zhang⁴
		¹Medical Imaging, Shanxi Medical University, Taiyuan, China, ²Department of Radiology, Shanxi Bethune Hospital, Taiyuan, China, ³MR Scientific Marketing, Diagnosis Imaging, Siemens Healthineers China, Beijing, China, ⁴MR Scientific Marketing, Diagnosis Imaging, Siemens Healthineers China, Shanghai, China
		Advanced DWI techniques, diffusion kurtosis imaging (DKI) and intravoxel incoherent motion (IVIM), can complement physiological information to morphological information that was obtained with conventional MRI This study investigated the application of DKI and IVIM in diagnosis of active sacroiliitis with ankylosing spondylitis.

Artificial Intelligence

Artificial Intelligence Applied to MSK MRI

Musculoskeletal

4045		Characterizing Knee Osteoarthritis Progression with Structural Phenotypes using MRI and Deep Learning
		Nikan K Namiri¹, Jinhee Lee¹, Bruno Astuto¹, Felix Liu¹, Rutwik Shah¹, Sharmila Majumdar¹, and Valentina Pedoia¹
		¹Department of Radiology and Biomedical Imaging and Center for Intelligent Imaging, University of California, San Francisco, San Francisco, CA, United States
		We built an end-to-end deep learning model to rapidly stratify knees into morphological phenotypes using a large, longitudinal cohort with knee osteoarthritis (OA). We examined associations of phenotypes with odds of concurrent OA and OA progression. Bone, meniscus/cartilage, and inflammatory phenotypes were strongly associated with current structural OA and symptomatic OA. Hypertrophy phenotype was only weakly associated with structural OA. Among those who did not have baseline OA, bone and meniscus/cartilage phenotypes were strongly associated with developing both structural and symptomatic OA in 48 months. Only bone phenotype increased risk of undergoing total knee replacement surgery within 96 months.

4046		Development of Deep Learning based Cartilage Segmentation at 3D knee MRI for the use of Biomarker of Osteoarthritis
		Jinwoo Han¹, Suk-Joo Hong¹, Zepa Yang¹, Woo Young Kang¹, Yoonmi Choi¹, Chang Ho Kang², Kyung-sik Ahn², Baek Hyun Kim³, and Euddeum Shim³
		¹Radiology, Korea University Guro Hospital, KUGH-MIDC, Seoul, Korea, Republic of, ²Korea University Anam Hospital, Seoul, Korea, Republic of, ³Korea University Ansan Hospital, Ansan, Korea, Republic of
		Cartilage loss is fundamental pathology of knee osteoarthritis (OA). Quantitative analysis of cartilage thickness and volume is very time consuming by manual measurement. We proposed development of deep learning based cartilage segmentation at three dimensional knee magnetic resonance images, which can measure thickness and volume of knee joint cartilage, automatically and accurately. To evaluate the performance, we used Dice Similarity Coefficient (DSC) respect to the manual segmentation, and visual inspection. The accuracy DSC values were higher than 0.9. We expect deep learning program can be useful in future study for knee joint osteoarthritis.

4047		Automation of Quantifying Axonal Loss in Patients with Peripheral Neuropathies through Deep Learning Derived Muscle Fat Fraction
		Yongsheng Chen¹, Daniel Moiseev¹, Wan Yee Kong¹, Alexandar Bezanovski¹, and Jun Li^1,2
		¹Department of Neurology, Wayne State University School of Medicine, Detroit, MI, United States, ²John D. Dingell VA Medical Center, Detroit, MI, United States
		Axonal loss determines the final disability in patients with peripheral neuropathies. Consequently, axonal loss results in intramuscular fat accumulation. Therefore, measuring muscle fat fraction through Dixon MRI has been a promising biomarker for monitoring disease progression. However, the responsiveness is yet to be improved, particularly in the early phase of the disease. In this study, we developed a deep learning-based method to automate the quantification of individual muscle fat fraction, which mitigates the laborious manual segmentations and enables the use of individual muscle fat fraction as outcome measures to track axonal loss in patients with neuropathies.

4048		Assessment of the potential of a Deep Learning Knee Segmentation and Anomaly Detection Tool in the clinical routine
		Laura Carretero¹, Pablo García-Polo¹, Suryanarayanan Kaushik ², Maggie Fung², Bruno Astuto^3,4, Rutwik Shah^3,4, Pablo F Damasceno^3,4, Valentina Pedoia^3,4, Sharmila Majumdar^3,4, and Mario Padrón⁵
		¹Global Research Organization, GE Healthcare, Madrid, Spain, ²GE Healthcare, Waukesha, WI, United States, ³Department of Radiology and Biomedical Imaging, UCSF, San Francisco, CA, United States, ⁴Center for Digital Health Innovation, UCSF, San Francisco, CA, United States, ⁵Department of Radiology, Clínica Cemtro, Madrid, Spain
		This study evaluates the clinical accuracy of a deep learning (DL)-based tool to segment articular cartilage and menisci on 50 knee MRI exams; detect lesions and stage its severity. An experienced MSK radiologist assessed independently the images for the presence of any lesions on the different compartments and checked the accuracy of its segmentation, resulting in no disagreement with the segmentation output in 92.8% of the compartments and correspondence in the detection of lesions in 75.94% of them. The shown results assessed the clinical potential of this tool and present a step forward into structured MSK imaging reports.

4049		Differentiation of Benign and Malignant Vertebral Fractures on Spine MRI Using ResNet Deep Learning Compared to Radiologists’ Reading
		Lee-Ren Yeh¹, Yang Zhang², Jeon-Hor Chen², An-Chi Wang³, JieYu Yang³, Peter Chang², Daniel Chow², and Min-Ying Su²
		¹Radiology, E-Da Hospital, Kaohsiung, Taiwan, ²University of California Irvine, Irvine, CA, United States, ³Radiology, Chi-Mei Medical Center, Tainan, Taiwan
		This study compared the reading of three radiologists with different level of experience, and also investigated the potential of deep learning to differentiate between benign and malignant vertebral fractures based on T1W and T2W MRI. The results showed that deep learning using ResNet50 achieved a satisfactory diagnostic accuracy of 92%, although inferior to 98% made by a senior MSK radiologist and 96% made by a R4 resident, much higher compared to 66% made by a R1 resident. The inferior performance of ResNet50 might be partly explained by the very limited information when only considering a small bounding box.

4050		Synovial Fluid Suppressed 3D T1ρ Mapping of Knee Cartilage using Deep Learning
		Can Wu^1,2 and Qi Peng³
		¹Department of Medical Physics, Memorial Sloan Kettering Cancer Center, New York, NY, United States, ²Philips Healthcare, Andover, MA, United States, ³Department of Radiology, Albert Einstein College of Medicine and Montefiore Medical Center, Bronx, NY, United States
		3D T1ρ mapping is a promising technique for quantitative assessment of biochemical changes in knee cartilage. However, synovial fluid, if not suppressed, may compromise T1ρ quantification, particularly in clinical conditions like osteoarthritis where cartilage is usually irregular and synovial fluid is increased. A long-T2-selective inversion approach can be used to suppress the synovial fluid signal at the cost of increased scan time by 50%. This study demonstrated that deep learning can be used to effectively eliminate synovial fluid from T1ρ data acquired without active fluid suppression, potentially leading to improved T1ρ quantification of knee cartilage accuracy without adding scan time.

4051		Deep CNNs with Physical Constraints for simultaneous Multi-tissue Segmentation and Quantification (MSQ-Net) of Knee from UTE MRIs
		Xing Lu¹, Yajun Ma¹, Saeed Jerban¹, Hyungseok Jang¹, Yanping Xue¹, Xiaodong Zhang¹, Mei Wu¹, Amilcare Gentili^1,2, Chun-nan Hsu³, Eric Y Chang^1,2, and Jiang Du¹
		¹Department of Radiology, University of California, San Diego, San Diego, CA, United States, ²Radiology Service, Veterans Affairs San Diego Healthcare System, San Diego, CA, United States, ³Department of Neurosciences, University of California, San Diego, San Diego, CA, United States
		In this study, we proposed end-to-end deep learning convolutional neural networks to perform simultaneous segmentation and quantification (MSQ-Net) on the knee without and with physical constraint networks (pcMSQ-Net). Both networks were trained and tested for the feasibility of simultaneous segmentation and quantitative evaluation of multiple knee joint tissues from 3D ultrashort echo time (UTE) magnetic resonance imaging. Results demonstrated the potential of MSQ-Net and pcMSQ-Net for fast and accurate UTE-MRI analysis of the knee, a “whole-organ” approach which is impossible with conventional clinical MRI.

4052		Deep-Learning Based Image Reconstruction for Lumbar Spine MRI at 3T: Clinical Feasibility
		Emma Bahroos¹, Misung Han¹, Cynthia Chin¹, David Shin², Javier Villanueva-Meyer¹, Thomas Link¹, Valentina Pedoia¹, and Sharmila Majumdar¹
		¹Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, ²Applications and Workflow, GE Healthcare, Menlo Park, CA, United States
		Lower back pain is one of the most common health problems, for which MRI is extensively used. Standard clinical, and fast acquisition images of lumbar spine were acquired for 18 patients. A (DL)-based image reconstruction was applied to the raw data of the fast images, with 25%, 50%, and 75% noise reduction factors. Evaluation of fast images with DL algorithm, for image quality, diagnostic capability, and SNR to standard images was conducted by three experienced radiologists. Our results show SNR improvement with higher noise reduction factor without a severe degradation in the ability to discern anatomical structures.

4053		Less is more: zero-shot detection and transfer learning for facet arthropathy localization and classification on lumbar spine MRIs
		Upasana Upadhyay Bharadwaj¹, Cynthia T Chin¹, Valentina Pedoia¹, and Sharmila Majumdar¹
		¹Radiology, University of California, San Francisco, San Francisco, CA, United States
		Lumbar facet arthropathy is frequently observed along with other degenerative changes of the spine in patients presenting with chronic low back pain. Deep learning has demonstrated unprecedented success in automated assessment of many spine degenerative changes, but heretofore not applied to facet arthropathy. This study presents binary classification of facet arthropathy (normal/mild vs moderate/severe) on T2-weighted axial MRI slices using a two-staged approach: zero-shot facet detection followed by classification. Our model achieves an AUC of 0.916 [0.911, 0.921] with sensitivity and specificity of 97.8% [97.4, 98.3] and 64.1% [63.1, 65.1], respectively and can potentially enhance the clinical workflow.

4054		DEMO: Deep MR Parametric Mapping using Unsupervised Multi-tasking Framework
		Jing Cheng¹, Yuanyuan Liu¹, Xin Liu¹, Hairong Zheng¹, Yanjie Zhu¹, and Dong Liang¹
		¹Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen, China
		In this work, we propose a novel deep learning-based framework DEMO for fast and robust MR parametric mapping. Different from current deep learning-based methods, DEMO trains the network in an unsupervised way. Specifically, a CS-based loss function is used in DEMO to avoid the necessity of using fully sampled k-space data as the label, and thus make it an unsupervised learning approach. DEMO reconstructs the parametric weighted images and generates the parametric map simultaneously, which enables multi-tasking learning. Experimental results show the promising performance of the proposed DEMO framework in quantitative MR T1ρ mapping.

4055		MRI image synthesis with a conditional generative adversarial network using patch pooling
		Bragi Sveinsson^1,2 and Matthew S Rosen^1,2,3
		¹Martinos Center, Massachusetts General Hospital, Boston, MA, United States, ²Harvard Medical School, Boston, MA, United States, ³Physics, Harvard University, Cambridge, MA, United States
		Deep learning networks allow the creation of new images based on a separate set of reference image data. This can be used to synthesize a specific MRI contrast from other image contrasts sharing the same anatomy. A particularly successful approach uses a conditional generative adversarial network with a patch-based discriminator, processing image patches of a fixed size. In this work, we investigate the benefits of using multiple patch sizes to improve image quality.

4056		Self-Supervised Deep Learning for Knee MRI Segmentation using Limited Labeled Training Datasets
		Jeffrey Dominic¹, Arjun Desai¹, Andrew Schmidt¹, Elka Rubin¹, Garry Gold¹, Brian Hargreaves¹, and Akshay Chaudhari¹
		¹Stanford University, Stanford, CA, United States
		Deep learning (DL)-based approaches have shown promise for automating medical image segmentation with high efficacy. However, current state-of-the-art DL supervised methods require large extents of labeled training images, which are difficult to curate at scale. In this work, we propose a self-supervised training scheme to reduce dependence on labeled data by pretraining networks in an unsupervised manner. We show that our method can improve segmentation performance, especially in the context of very limited data scenarios (only 10-25% scans available) and can achieve or surpass the accuracy of state-of-the-art supervised networks with approximately 50% fewer labeled scans.

4057		Deep Learning Improves Detection of Anterior Cruciate Ligament- and Meniscus Tear Detection in Knee MRI
		Firas Khader¹, Gustav Müller-Franzes¹, Johannes Stegmaier², Martin Pixberg³, Jonas Müller-Hübenthal³, Christiane Kuhl ¹, Sven Nebelung⁴, and Daniel Truhn¹
		¹Department of Diagnostic and Interventional Radiology, Aachen University Hospital, Aachen, Germany, ²Institute of Imaging and Computer Vision, RWTH Aachen University, Aachen, Germany, ³Praxis im Köln Triangle, Cologne, Germany, ⁴Department of Diagnostic and Interventional Radiology, Düsseldorf University Hospital, Dusseldorf, Germany
		In this study we aimed to analyze the capability of neural networks to accurately diagnose the presence of ACL and meniscus tears in our in-house dataset comprised of 3887 manually annotated knee MRI exams. To this end we trained the MRNet architecture on a varying number of training exams that included proton density-weighted axial, sagittal and coronal planes for each knee exam. Additionally, we compared the performance of the architecture when trained on expert vs non-expert annotations. This study demonstrates that while our neural network benefits from a larger dataset, expert annotations do not considerably improve the performance.

4058		Fully automatic detection and voxel-wise mapping of vertebral body Modic changes using deep convolutional neural networks
		Kenneth T Gao^1,2,3, Radhika Tibrewala^1,2, Madeline Hess^1,2, Upasana Bharadwaj^1,2, Gaurav Inamdar^1,2, Cynthia T Chin¹, Valentina Pedoia^1,2, and Sharmila Majumdar^1,2
		¹Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, ²Center for Intelligent Imaging, Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, ³University of California, Berkeley-University of California San Francisco Graduate Program in Bioengineering, San Francisco, CA, United States
		Modic changes are common degenerative lesions seen in spinal MRI and are strongly linked to lower back pain. However, detection of Modic changes suffers from poor inter-operator and inter-scanner reliabilities. We present a fully automatic, quantitative model that leverages deep learning and signal-based clustering for mapping Modic changes from clinically acquired MRI. The model achieves an identification rate of 85.7% and substantial agreement with radiologists. More importantly, the mapping technique classifies detected lesions on a voxel-wise basis, allowing for assessment of sensitive, local pathologies.

4059		Towards Clinical Translation of Fully Automatic Segmentation and 3D Biomarker Extraction of Lumbar Spine MRI
		Madeline Hess¹, Kenneth Gao¹, Radhika Tibrewala¹, Gaurav Inamdar¹, Upasana Bharadwaj¹, Cynthia Chin¹, Valentina Pedoia¹, and Sharmila Majumdar¹
		¹Center for Intelligent Imaging, University of California, San Francisco, San Francisco, CA, United States
		Lumbar spine segmentation serves as an important first step for automated disease classification and monitoring, but manual segmentation is costly and time consuming. We present a deep learning-based pipeline to automatically segment the vertebral bodies, intervertebral discs, and paraspinal muscles in the lumbar spine. We leverage the results of this method to quickly and accurately extract disc height with a mean absolute error of 2.09 mm, muscle CSA with mean absolute errors of less than 1.46 cm2, and muscle centroid position with a mean absolute error of less than 7.23mm.

4060		A pipeline combining deep learning and radiomics to automatically identify chronic lateral ankle instability from FS-PD MRI
		Yibo Dan¹, Hongyue Tao², Chengxiu Zhang¹, Chenglong Wang¹, Yida Wang¹, Shuang Chen², and Guang Yang¹
		¹Shanghai Key Laboratory of Magnetic Resonance, East China Normal University, shanghai, China, ²Department of Radiology, Huashan Hospital, Fudan University, shanghai, China
		The naked eye can only recognize the morphological changes of cartilage and subchondral bone on conventional MRI, but cannot recognize the subtle changes in their internal structure. The aim is to use radiomics to evaluate the cartilage and subchondral bone changes in patients with chronical ankle joint instability (CAI) on conventional MRI images¹. We built a pipeline to automatically identify CAI from FS-PD images. The pipeline automatically segmented cartilage regions and subchondral bone (5mm) regions, then used SVM based on radiomics features extracted from these regions for classification. In the test dataset, the proposed model achieved an AUC of 0.965.

4061		Deep Learning Reconstruction of 3D Zero Echo Time Magnetic Resonance Images for the Creation of 3D Printed Anatomic Models
		Nicole Wake^1,2, Stephanie Shamir¹, Beverly Thornhill¹, Nogah Haramati¹, Graeme McKinnon³, Mathias Engstrom⁴, Florian Wiesinger⁴, Michael Carl⁵, Fraser Robb⁶, and Maggie Fung⁷
		¹Department of Radiology, Montefiore Medical Center, Bronx, NY, United States, ²Center for Advanced Imaging Innovation and Research, Department of Radiology, NYU Langone Health, New York, NY, United States, ³GE Healthcare, Waukesha, WI, United States, ⁴GE Healthcare, Munich, Germany, ⁵GE Healthcare, San Diego, CA, United States, ⁶GE Healthcare, Aurora, OH, United States, ⁷GE Healthcare, New York, NY, United States
		Patient-specific three-dimensional (3D) printed anatomic models are valuable clinical tools which are generally created from computed tomography (CT). However, magnetic resonance imaging (MRI) is an attractive alternative, since it offers exquisite soft-tissue characterization and flexible image contrast mechanisms while avoiding the use of ionizing radiation. The purpose of this study was to evaluate the image quality and assess the feasibility of creating 3D printed models using a 3D Zero echo time (ZTE) MR images which were reconstructed with a deep learning reconstruction method.

4062		Identification of Bone Marrow Lesions on Magnetic Resonance Imaging with Weakly Supervised Deep Learning
		Jiaping Hu¹, Zhao Wang², Lijie Zhong¹, Keyan Yu¹, Yanjun Chen¹, Yingjie Mei³, Qi Dou⁴, and Xiaodong Zhang¹
		¹Department of Medical Imaging, The Third Affiliated Hospital of Southern Medical University, Guangzhou, China, ²College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China, ³China International Center, Philips Healthcare, Guangzhou, China, ⁴Department of Computer Science & Engineering, The Chinese University of Hong Kong, Hong Kong, China
		The presence of a bone marrow lesion is associated with incident and progressive knee osteoarthritis (KOA) and joint replacement. Since the ill-defined boundary and various signal strength, identification of bone marrow lesions (BMLs) requires professional diagnostic ability and is subjective. Therefore, we utilize a model to assess whether there exists BMLs in every subregion and their severity on 3D-dual echo steady state (DESS) images according to MRI Osteoarthritis Knee Score (MOAKS). The initiatory results showed that deep learning framework performed well on discrimination of BMLs with good reproducibility.

4063		Retrospective Contrast Tuning from a Single T1-weighted Image Using Deep Learning
		Yan Wu¹, Yajun Ma², Jiang Du², and Lei Xing¹
		¹Stanford University, Palo Alto, CA, United States, ²University of California San Diego, San Diego, CA, United States
		While versatile soft tissue contrasts are achievable in MRI, contrast attainable from each scan is predetermined by the imaging protocol. A retrospective tuning of contrast will provide an opportunity to normalize MRI data for radiomics analysis. In this study, we present a new paradigm to obtain a spectrum of contrasts from a single T1-weighted image. Using deep learning, T1 map, proton density map, and B1 map are predicted from every T1-weighted image, and new contrasts can be obtained with the application of Bloch equations. The method has been validated in knee MRI with high accuracy achieved.

4064		Feasibility of Femoral Cartilage Lesion Classification on Clinical MRIs using Deep Learning
		Mingrui Yang¹, Ceylan Colak¹, Mercan Aslan¹, Sibaji Gaj¹, Morgan Jones¹, Carl Winalski¹, Naveen Subhas¹, and Xiaojuan Li¹
		¹Cleveland Clinic, Cleveland, OH, United States
		Early diagnosis and effective detection of cartilage degeneration is an important factor for osteoarthritis prevention and treatment, which are still challenging in routine clinical practice, resulting in poor patient treatment and management plans. The purpose of this study is to assess the feasibility of building an automatic femoral cartilage lesion classification pipeline for heterogenous clinical routine MR scans by combining deep learning segmentation and classification models together.

Bone Tendon Inflammation

Artificial Intelligence Applied to MSK MRI

Musculoskeletal

4065		Discussion the Value of ADC in Bone Contusions and Bone Marrow Lesions in Osteoarthritis of Knee Joint
		Weixin He¹, Qi Zeng¹, Ziwei Zhang¹, Xia Zhu¹, Zhaoshu Huang², Lisha Nie³, and Lingling Song¹
		¹Affiliated Hospital of Guizhou Medical University, Guiyang, China, ²592159673@qq.com, Guiyang, China, ³GE Healthcare, MR Research China, Beijing, China
		The current study aims to investigate the application value of ADC in Bone Contusions and Bone Marrow Lesions in Osteoarthritis of Knee Joint. It was concluded that ADC value has the potential value to differentiate the two kinds of lesions.

4066		Pilot contrast-free MRI reveals significantly impaired calf skeletal muscle perfusion in diabetes with incompressible peripheral arteries
		Jie Zheng¹, Sara Gharabaghi², Ran Li¹, Yongsheng Chen³, Hongyu An¹, E Mark Haacke², Mohamed A Zayed¹, and Mary K Hastings¹
		¹Washington University in St. Louis, St. Louis, MO, United States, ²MR Innovations Inc, Bingham Farms, MI, United States, ³Wayne State University, Detroit, MI, United States
		The purpose of this study is to investigate whether individuals with diabetes mellitus and incompressible arteries will have lower skeletal muscle microcirculation during a moderate isometric exercise. Healthy volunteers, diabetes with normal artery, and diabetes with incompressible arteries underwent MR angiography and calcification imaging, and skeletal muscle microcirculation imaging at rest and during an isometric contraction exercise. Significantly lower perfusion reserve and its association with higher ABI were observed in diabetes with incompressible arteries. Despite lack of apparent arterial stenosis, calcification can be readily visualized in 2 patients.

4067		Acceleration of high-resolution proximal femur MRI using compressive sensing and sparsity in a retrospective study
		Brian-Tinh Duc Vu^1,2, Brandon Jones^1,2, Winnie Xu², Gregory Chang³, and Chamith Rajapakse^2,4
		¹Bioengineering, University of Pennsylvania, Philadelphia, PA, United States, ²Radiology, University of Pennsylvania, Philadelphia, PA, United States, ³Radiology, Center for Biomedical Imaging, New York University, New York, NY, United States, ⁴Orthopaedic Surgery, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, United States
		Retrospective compressive sensing techniques demonstrate promise accelerating acquisition of high-resolution images of the femur while maintaining sufficient image quality to assess fracture risk. While the trabecular bone microstructure was preserved at an undersampling rate of 30%, lower sampling rates of 10% and 5% exhibited visually apparent artifacts and image degradation. Similarly, bone stiffness at 30% resembled fully sampled data but the error increased as sampling rate decreased. Nevertheless, the results show that compressive sensing is a promising candidate for accelerating the acquisition rate, and further prospective studies are needed to further validate this finding.

4068		3T MRI Distribution of Textural Features in Bone Marrow for Osteoporosis.
		Anmol Monga¹, Dimitri Martel¹, Stephen Honig², and Gregory Chang¹
		¹Department of Radiology, NYU Langone Health, New York, NY, United States, ²Osteoporosis Center, Hospital for Joint Disease, NYU Langone Health, New York, NY, United States
		In this preliminary study we aim to analyze the relationship of first-order textural feature of Fat, water, Fat Fraction maps, clinical features (Age, height, and weight) and BMD (Hip, Spine and Femoral Neck). Radiomic features calculated on Fat parametric maps explains variability in Bone Mineral Density to higher extend than water and fat fraction parametric maps.

4069		Significantly reduced collagen and increased water in tibia of patients with osteopenia and osteoporosis detected with ultrashort echo time MRI
		Saeed Jerban¹, Yajun Ma¹, Zhao Wei¹, Meghan Shen¹, Amir Masoud Afsahi¹, Zubiad Ibrahim¹, Alecio Lombardi^1,2, Douglas G Chang³, Eric Y Chang^1,2, and Jiang Du¹
		¹Radiology, University of California, San Digeo, La Jolla, CA, United States, ²Radiology Service, VA San Diego Healthcare System, San Diego, CA, United States, ³Orthopaedic Surgery, University of California, San Digeo, La Jolla, CA, United States
		Assessment of water and collagen content in bone is missing from DEXA evaluation, in standard osteoporosis diagnosis. Bone signals in ultrashort echo time MRI (UTE-MRI) and inversion recovery UTE-MRI against a known external reference signal were used to measure total, bound, and pore water proton densities. Macromolecular proton density was estimated by multiplying total water proton density with macromolecular fraction derived from UTE magnetization transfer (UTE-MT) modeling. The UTE-evaluated MRI measures demonstrated significantly reduced collagen and increased water content in the tibia of patients with osteoporosis and osteopenia compared with healthy subjects. Hip T-scores showed significant correlation with UTE-MRI measures.

4070		A comparative study on MRI-based radiomics model selection for a high-risk cytogenetics prediction in multiple myeloma
		jianfang liu¹ and huishu yuan¹
		¹peking university third hospital, beijing, China
		We aim to develop a radiomics model based on MRI to predict high-risk cytogenetics (HRC) status in patients with multiple myeloma (MM) and identify optimal machine learning methods. We retrospectively analyzed 89 patients’ (37 HRC, 52 non-HRC) radiomics features extracted from fat suppression T2W and TIW image. The following classification methods, including, support vector machine, random forest, logistic regression (LR) and decision tree were used to construct radiomics models. LR model showed the highest performance with an AUC of 0.82 ± 0.02. Radiomics model based on LR classifier can be used to predict HRC status in patients with MM effectively.

4071		Impact of sustained Synovitis on Knee Joint Structural Degeneration: 4-Year MRI Data from the Osteoarthritis Initiative
		Sara Ramezanpour¹, Thanat Kanthawang ^1,2, John Lynch ³, Thomas M Link ¹, and Gabby B Joseph¹
		¹Department of Radiology and Biomedical Imaging, University of California, San Francisco, San Francisco, CA, United States, ²Department of Radiology, Faculty of Medicine, Chiang Mai University, Thailand, Chiang Mai, Thailand, ³Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA, United States
		Synovial inflammation is a well-known risk factor of osteoarthritis (OA). Several semi-quantitative scores have been developed to grade synovitis using non-contrast enhanced MRI studies of the knee. However, the role of inflammatory processes in the progression of knee OA is yet to be fully understood. One hundred and eighty individuals with right knee MRIs at baseline, 2 and 4 years were studied using MRI-based semi-quantitative synovitis scores measuring presence/absence as well as severity of synovitis Subjects with sustained synovitis had greater progression of meniscus, bone marrow and cartilage abnormalities compared to controls without sustained synovitis.

4072		AcidoCEST MRI pH is strongly correlated with GLUT1 immunohistochemistry in multiple myeloma
		Alecio Lombardi^1,2, Jonathan Wong^1,2, Rachel High^1,2, Ya-Jun Ma², Adam Searleman², Saeed Jerban², Qingbo Tang^1,2, Jiang Du^1,2, Patrick Frost^3,4, and Eric Y. Chang^1,2
		¹Radiology Service, Veterans Affairs San Diego Healthcare System, San Diego, CA, United States, ²Radiology, University of California, San Diego, CA, United States, ³Greater Los Angeles Veteran Administration Healthcare System, Los Angeles, CA, United States, ⁴University of California, Los Angeles, CA, United States
		Multiple myeloma (MM) is a malignant plasma cell disease. Adaptive responses to hypoxia may be an essential element in its progression. The purpose of this study was to determine the feasibility of acidoCEST MRI for pHe measurement on a mouse model of MM with comparison with GLUT1 staining.

4073		Diagnostic performance of zero echo time imaging and T1-weighted fast spin echo on sacroiliac joint bone erosions using CT as the gold standard
		Yitong Li¹, Shuang Hu¹, Weiyin Vivian Liu², and Xiaoming Li¹
		¹Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China, ²MR Research, GE Healthcare, Beijing, China
		Detection of sacroiliac joint (SIJ) bone erosions may improve diagnostic performance of axial spondyloarthritis (axSpA) as CT is the gold standard. Zero echo time (ZTE) MRI technique can acqire CT-like bone contrast and has been applied to bone imaging of many other body parts. Our research explored the ability of ZTE to detect SIJ bone erosions and conduct comparison of ZTE and conventional T1-weighted fast spin echo (T1 FSE). We found that ZTE had superior detection performance to T1 FSE and might be an effective supplement to routine MRI protocol in SIJs.

4074		Evaluation of the risk of osteoporosis in diabetic patients by IDEAL-IQ
		Yu Song¹, Qingwei Song², Yingkun Guo¹, Gang Ning¹, and Xuesheng Li¹
		¹Department of Radiology, West China Second University Hospital, Sichuan University, Chengdu, China, ²the First Affiliated Hospital of Dalian Medical University, Dalian, China
		Osteoporosis is a progressive metabolic bone lesion characterized by a decrease in bone mass per unit volume, which results in increased risk of bone fractures. Noninvasive and reliable assessment of osteoporosis is essential for clinical diagnosis and treatment. Studies have reported that precise measurement of fat content in the bone marrow is a biomarker for quantifying osteoporosis. IDEAL-IQ is a method used for quantitative measurement of fat fraction and iron content related parameters by multi-echo acquisition.

4075		Detection of Sacroiliac Joint Lesions in Axial Spondyloarthritis:Utility of Synthetic MRI
		Ke Zhang¹ and Guobin Hong¹
		¹Radiology, the Fifth Affiliated Hospital, Sun Yat-sen University, Zhuhai, China
		The current diagnosis of the sacroiliac joint lesions with axSpA is most concentrated on qualitative or semi-quantitative methods.There is an urgent need for a quantitative method that can objectively and accurately evaluate the severity of the disease.Synthetic MRI(MAGiC) can generate multiple contrast images and quantitative maps simultaneously based on the same scan.In this study the synthetic MRI can achieve similar qualitative diagnostic performance in detection of sacroiliac joint lesions compared with conventional MRI.And it could be used for distinguishing BME and fat metaplasia.

4076		MRI and CT in an Ancient Child Mummy: Contrast Combination to Increase Tissue Differentiation
		Agazi Samuel Tesfai¹, Johannes Fischer¹, Ali Caglar Özen^1,2, Patrick Eppenberger³, Lena Öhrström³, Frank Rühli³, Ute Ludwig¹, and Michael Bock¹
		¹Dept. of Radiology, Medical Physics, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany, ²German Consortium for Translational Cancer Research Partner Site Freiburg, German Cancer Research Center (DKFZ), Heidelberg, Germany, ³Institute of Evolutionary Medicine, Faculty of Medicine, University of Zurich, Zurich, Switzerland
		MR images of a child mummy were acquired on a clinical 3T system using a dedicated RF coil with optimized RF switching hardware and 3D UTE sequence. In addition, dual-energy CT images were sampled and co-registered to compare MRI signal intensities and T2* relaxation times with CT Hounsfield Units and effective atomic numbers in bone, soft tissue and embalming material.

4077		Associations between Bone Turnover Biomarkers and MR-based Knee composition and WORMS scores:cross-sectional
		WANG BIN¹, TAN HUI¹, HE Taiping¹, YU Nan¹, and WANG Shaoyu²
		¹Shaanxi University of Chinese Medicine, SHAANXI, China, ²Siemens Healthineers, shanghai, China
		We attempted to explore the relationship between bone markers (including bone formation and bone absorption) and cartilage, meniscus morphology, and WORMS socres in patients with osteoarthritis.

4078		MRI-based radiomic features and machine learning for differentiating myelodysplastic syndrome and aplastic anemia
		Miyuki Takasu¹, Makoto Iida¹, Yasutaka Baba², Yuji Akiyama¹, Yuji Takahashi¹, Takashi Abe³, and Kazuo Awai¹
		¹Department of Diagnostic Radiology, Hiroshima University Hospital, Hiroshima, Japan, ²Department of Radiology, International Medical Center, Saitama Medical University, Saitama, Japan, ³Department of Radiology, Nagoya University Hospital, Aichi, Japan, Nagoya, Japan
		We assessed the feasibility of a method of radiomic analysis based on machine learning (ML) and lumbar MRI to differentiate between MDS and aplastic anemia (AA). Regions of interest were drawn in the L3 vertebral body on the mid-sagittal images of sagittal T1-weighted and STIR images of patients with MDS (n=62) or AA (n=78) from six institutions. The model of ML with logistic regression resulted in the best performance for differentiating MDS from AA when using T1-weighted images. The model was not predictive for STIR or concatenated images. The radiomics-based ML model enabled the differentiation of MDS and AA.

4079		Evaluation of Synovitis of Hand in Patients with Rheumatoid Arthritis Using Diffusion Kurtosis Imaging: Initial Findings
		Kaifang Liu¹, Jie Meng¹, and Zhengyang Zhou¹
		¹Departments of Radiology, Nanjing Drum Tower Hospital Clinical College of Nanjing Medical University, Nanjing, China
		Thirty patients with RA and 10 patients suspected of RA were enrolled to explore the application value of DKI in the noninvasive identification of synovitis in hand arthritis. The suspected synovitis or joint effusion was scored on a scale of 0 (joint effusion) to 3 (mild, moderate, severe synovitis), referring to RAMRIS (RA-MRI-Scoring) system. ADC, D, and K from DKI were recorded and compared. There were significant differences in ADC, D, and K values among different enhancing degree scores. The diagnostic performance provided by the D values is similar to the ADC value and higher than the K value.

4080		A Comparison of Synovitis Severity in the Knee Assessed Using Contrast-Enhanced MRI and FDG-PET
		Jacob Thoenen¹, James W. MacKay^2,3, Kathryn J. Stevens¹, Tom D. Turmezei⁴, Akshay Chaudhari¹, Lauren E. Watkins¹, Brian A. Hargreaves¹, Garry E. Gold¹, and Feliks Kogan¹
		¹Department of Radiology, Stanford University, Stanford, CA, United States, ²Department of Radiology, University of Cambridge, Cambridge, United Kingdom, ³Norwich Medical School, University of East Anglia, Norwich, United Kingdom, ⁴Department of Radiology, Norfolk and Norwich University Hospital, Norwich, United Kingdom
		T1-weighted contrast-enhanced MRI (CE-MRI) is used to evaluate synovitis, but fluctuations in severity of synovial inflammation may be slow to manifest as volumetric changes seen on CE-MRI. We compared metabolic activity using FDG-PET to CE-MRI synovitis gradings, MOAKS effusion- and Hoffa-synovitis gradings, and the transfer constant K^trans in the overall synovial region and four subregions. Moderate to very strong correlation was seen between maximum FDG uptake (SUV_max) and corresponding synovitis grades, MOAKS effusion synovitis, and corresponding regional median K^trans both within the entire synovium and in four synovial subregions, with the exception of median K^trans at one subregion.

4081		A preliminary study on the effect of compressed SENSE with multiple acceleration factors on knee examination
		Nan Zhang¹, Qingwei Song², Ailian Liu², Renwang Pu², Haonan Zhang², Jiazheng Wang³, and Liangjie Lin³
		¹The First Affilliated Hospital of Dalian Medical University, Dalian, China, ²The First Affiliated Hospital of Dalian Medical University, Dalian, China, ³Philips Healthcare, Beijing, China, Beijing, China
		2D scans remain as standard for routine examination of joint lesions due to the short scan time. This study aims to explore feasibility of compressed SENSE with different acceleration factors in 3D high-resolution PD weighted imaging of knee joint.

4082		Diffusion MRI of THAs for the Classification of Synovial Reactions
		Madeleine A. Gao¹, Ek T. Tan¹, John Neri¹, Bin Lin¹, Alissa J. Burge¹, Hollis G. Potter¹, Kevin M. Koch², and Matthew F. Koff¹
		¹Radiology and Imaging, Hospital of Special Surgery, New York, NY, United States, ²Medical College of Wisconsin, Milwaukee, WI, United States
		The classification of synovial reactions caused by adverse local tissue reactions or prosthetic joint infections in total hip arthroplasty patients has remained a challenge due to the in-plane and through-plane distortions caused by metallic components. MAVRIC-based diffusion weighted imaging (DWI) is an innovative MRI technique sensitive to microstructural fluid changes and utilizes apparent diffusion coefficient values for the differentiation of diseased and non-diseased tissues. This study evaluated the effectiveness of MAVRIC-based DWI and T₂ mapping in differentiating between types of synovial reactions. Our DWI ADC biomarker showed promise in differentiating between normal and abnormal synovial reaction subtypes within THA patients.

4083		Clinical value of MAVRIC-SL MRI for the assessment of periprosthetic joint infection
		Tsutomu Inaoka¹, Masayuki Sugeta¹, Noriko Kitamura¹, Tomoya Nakatsuka¹, Rumiko Ishikawa¹, Takamitsu Uchi¹, Rui Iwata¹, Akinori Yamamoto¹, Hisanori Tomobe¹, Ryosuke Sakai¹, Hidetoshi Yamana¹, Shusuke Kasuya¹, and Hitoshi Terada¹
		¹Radiology, Toho University Sakura Medical Center, Sakura, Japan
		MAVRIC-SL enabled to improve the image quality by decreasing susceptibility artifacts and the assessment of periprosthetic joint infection after joint arthroplasty. Joint effusion, soft-tissue fluid collection, and soft-tissue edema were suggestive of periprosthetic joint infection. Soft-tissue fluid collection and soft-tissue edema were indicative of therapeutic surgical intervention for periprosthetic joint infection. Using MAVRIC-SL, MRI can become a useful tool to suggest periprosthetic joint infection in arthroplasty patients and to determine the management of periprosthetic joint infection.

4084		Diagnostic value of intro-voxel incoherent movement (IVIM) for infrapatellar fat pad high signal intensity in patient with osteoarthritis
		Hui Tan¹, Bin Wang¹, Wulin Kang¹, Nan Yu¹, Yong Yu¹, Shaoyu Wang², Yue Li¹, and Tuona Di¹
		¹Shaanxi University of Chinese Medicine, Xianyang, China, ²Siemens Healthineers, Shanghai, China
		This study aimed to investigate the role of IVIM in assessing infrapatellar fat pad high signal intensity in patients with osteoarthritis. Totally 43 patients with mild to moderate (K/L score = 1,2 and 3) KOA were included. The results show that D value, VAS, WOMAC, K/L score with T2FS-hyperintense regions were significantly higher than those without T2FS-hyperintense regions. Furthermore, D value was significantly positive associated with VAS and K/L score.

Quantitative MRI I

Quantitative MSK MRI

Musculoskeletal

4203		MIXTURE: A novel sequence for simultaneous morphological and quantitative imaging based on multi-interleaved 3D turbo-spin echo MRI
		Masami Yoneyama¹, Takayuki Sakai², Shuo Zhang³, Daichi Murayama², Hajime Yokota⁴, Yansong Zhao⁵, Shinji Saruya⁶, Masashi Suzuki⁶, Atsuya Watanabe^7,8, Mamoru Niitsu⁶, and Marc Van Cauteren⁹
		¹Philips Japan, Tokyo, Japan, ²Department of Radiology, Eastern Chiba Medical Center, Chiba, Japan, ³Philips Healthcare, Hamburg, Germany, ⁴Department of Diagnostic Radiology and Radiation Oncology, Graduate School of Medicine, Chiba University, Chiba, Japan, ⁵Philips Healthcare, Cleveland, OH, United States, ⁶Department of Radiology, Saitama Medical University, Saitama, Japan, ⁷Department of Orthopaedic Surgery, Eastern Chiba Medical Center, Chiba, Japan, ⁸General Medical Services, Graduate School of Medicine, Chiba University, Chiba, Japan, ⁹Philips Healthcare, Best, Netherlands
		3D high-resolution magnetic resonance imaging (MRI) is clinically essential for an accurate assessment of subtle pathophysiological changes in different anatomies. However, long exam time remains a practical challenge that requires additional scans for either complementary quantitative parametric mapping or morphological information for clinical diagnosis. In this study, we introduce a novel method termed Multi-Interleaved X-prepared TSE with inTUitive RElaxometry (MIXTURE) for simultaneous morphological and quantitative imaging with volumetric isotropic resolution within one single scan and demonstrate the approach in multi-contrast quantitative knee imaging with direct T2- or T1ρ-mapping.

4204		Annular tears in the intervertebral disc can be detected and highlighted using conventional MRI, texture analysis and neural networks
		Christian Waldenberg¹, Stefanie Eriksson¹, Hanna Hebelka², Helena Brisby³, and Kerstin Magdalena Lagerstrand¹
		¹Department of Radiation Physics, University of Gothenburg, Gothenburg, Sweden, ²Department of Radiology, University of Gothenburg, Gothenburg, Sweden, ³Department of Orthopaedics, University of Gothenburg, Gothenburg, Sweden
		Annular tears of intervertebral discs (IVDs) are associated with ingrowth of nerve endings with the potential to cause discogenic pain. We propose a novel non-invasive workflow for detecting annular tears and determining its location based on conventional magnetic resonance imaging (MRI). 124 IVDs in 44 patients with low back pain were included. Based on MRI, texture analysis and neural networks, a classification algorithm and workflow for determining the presence and position of annular tears were proposed. Excellent classification sensitivity of 100% and specificity of 93% were reached. The position of the tears was correctly determined in 67% of the IVDs.

4205		Lumbar Fatty Acid Composition (FAC) Measurement in Osteoporosis Patients Using MRI
		Lina Lin¹, Lutian Bai¹, Qun Cheng², Manuel Schneider³, Marcel Dominik Nickel³, Caixia Fu⁴, Guangwu Lin¹, and Shihong Li¹
		¹Department of Radiology, Huadong Hospital, Fudan University, Shanghai, China, ²Department of Osteoporosis, Huadong Hospital, Fudan University, Shanghai, China, ³MR Application Predevelopment, Siemens Healthcare GmbH, Erlangen, Germany, ⁴Application Development, Siemens Shenzhen Magnetic Resonance Ltd, Shenzhen, China
		This study is to explore the application value of fatty acid composition (FAC) measurements in the diagnosis of osteoporosis. 39 subjects who underwent dual-energy X-ray absorptiometry (DXA) and lumbar magnetic resonance imaging on a 3T scanner were included. We found that FF value was negatively correlated with T value (P < 0.05, r = -0.371), and SF value was positively correlated with T value (P < 0.05, r = 0.399). The PUF value of the osteopenia group was significantly higher than that of the osteoporosis group（P < 0.05）and normal bone mass group（P < 0.05）.

4206		Association of cartilaginous endplate changes in degenerative intervertebral discs and vertebral marrow fat: a quantitative MRI study
		Yayun Ji^1,2, Jianqiang Fang², Binyu Zhang², Siyuan Mi², Weiyin Vivian Liu³, Weian Zhao², and Liheng Ma¹
		¹The First Affiliated Hospital of Guangdong Pharmaceutical University, Guangzhou, China, ²Xianyang Central Hospital, Xianyang, China, ³MR research, GE Healthcare, Beijing, China
		This study aims to investigate if advanced UTE T2* Mapping MRI quantitative sequence could sensitively assess the early changes in the biochemical components of the cartilage endplate. This might explain the pathological mechanism of cartilage endplate injury, that is, bone marrow fat content increased in vertebral body of degenerated intervertebral disc. To sum up, UTE T2* Mapping could a potential imaging tool to early diagnose cartilage endplate changes and assist clinical prevention of lumbar intervertebral disc degeneration at an earlier stage.

4207		3D-ultrashort echo time imaging evaluation of the sacroiliac joint in patients with ankylosing spondylitis
		Cui Ren¹, Qing Li², Stefan Sommer^3,4, Qiao Zhu¹, and Huishu Yuan¹
		¹Radiology, Peking University Third Hospital, Beijing, China, ²MR Collaborations, Siemens Healthcare Ltd, Shanghai, China, ³Siemens Healthcare, Zurich, Switzerland, ⁴Swiss Center for Musculoskeletal Imaging (SCMI), Zurich, Switzerland
		The aim of this study was to evaluate the performance of a 3D-UTE sequence in the detection of sacroiliac joint cartilage and bone cortex abnormalities in patients with clinically confirmed ankylosing spondylitis. T2* derived from 3D-UTE displayed potential as a quantitative measure to diagnose ankylosing spondylitis.

4208		A Combined Solid-State 1H and 31P Magnetic Resonance Imaging to Assess Bone Mineral and Matrix Densities in Rat bones
		Victor Babu Kassey^1,2,3,4, Matthias Walle¹, Jonathan Egan¹, Diana Yeritsyan¹, Yaotang Wu^2,3,4, Brian D Snyder^2,4, Edward Rodriguez^1,2, Jerome Ackerman^2,5, and Ara Nazarian^1,2,4
		¹Carl J. Shapiro Department of Orthopaedic Surgery, Beth Israel Deaconess Medical Center, Boston, MA, United States, ²Harvard Medical School, Boston, MA, United States, ³Massachusetts General Hospital, Charlestown, MA, United States, ⁴Department of Orthopaedic Surgery, Boston Children's Hospital, Boston, MA, United States, ⁵Department of Radiology, Massachusetts General Hospital, Charlestown, MA, United States
		Bone matrix and mineral densities (BMD) are important parameters to identify bone diseases such as osteoporosis and osteomalacia. Micro-CT, CT, and DXA scans provide bone mineral densities but not bone matrix. In this study, a non-invasive, radiation-free, and clinically proven combined 1H/31P MRI method was developed to measure bone matrix and mineral densities from rat bones from the same volume-of-interest sequentially in a single session. A custom-designed home-made double-tuned single volume coil was designed for 7T, and 1H/31P ZTE rat bone images were obtained, auto-registered, bone matrix and mineral densities were computed quantitatively, and osteoporosis and osteomalacia were successfully identified.

4209		Monitoring changes of knee with Amateur Marathon athletes using Synthetic Magnetic Resonance Imaging ：A preliminary study
		Yijie Fang¹, Wenjun Yu¹, Long Qian², and Shaolin Li¹
		¹The Fifth Affiliated Hospital of Sun Yat-sen University, Zhuhai, China, ²MR Research, GE Healthcare, Beijing, China
		To explore the value of Synthetic magnetic resonance imaging (Sy MRI) in the quantitative monitoring of knee joint structural changes and cartilage of amateur marathon runners before and after the whole marathon. Twenty-six knee joints of healthy volunteers who took part in the marathon were scanned by Sy MRI technology one week before the race and within 48 hours after the race. The values of T1, T2 and PD of knee cartilage in whole and different regions before competition were higher than those of 48 hours after the marathon, and the difference was statistically significant

4210		Glycosaminoglycan detection in OA patients using AC-iTIP at 3.0T: A preliminary study
		Baiyan Jiang¹, Ki-wai, Kevin Ho², Xiao Fan¹, Jian Hou¹, Chun-man Lawrence Lau², James Griffith¹, and Weitian Chen¹
		¹Imaging & Interventional Radiology, The Chinese University of Hong Kong, Sha Tin, Hong Kong, ²Orthopaedics & Traumatology, The Chinese University of Hong Kong, Sha Tin, Hong Kong
		The measurement of R1rho (1/T1rho) spectrum and its asymmetry have several advantages over conventional CEST MTR asymmetry in probing chemical exchange effect. Previous studies have reported that AC-iTIP is a robust approach to obtain R1rho-spectrum and R1rho asymmetry. However, an accurate B0 map is required to perform R1rho asymmetry. In this work, we fit the entire R1rho spectrum to a mathematical expression and directly acquire the pool population parameter pb. A preliminary in vivo scan on 4 osteoarthritis patients was performed and AC-iTIP shows promising results in probing GAG content in knee cartilage at 3.0T.

4211		Measuring the effect of mechanical loading on deep and superficial cartilage using quantitative UTE MRI
		Hanqi Wang¹, Qing Li², Stefan Sommer^3,4, and Yong Lu¹
		¹Ruijin Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China, ²MR Collaboration, Siemens Healthcare Ltd., Shanghai, China, Shanghai, China, ³Siemens Healthcare, Zurich, Switzerland, Zurich, Switzerland, ⁴Swiss Center for Musculoskeletal Imaging (SCMI), Balgrist Campus, Zurich, Switzerland, Zurich, Switzerland
		UTE sequences were utilized to quantitatively assess deep and superficial cartilage. Deep cartilage T2* values demonstrated a significant difference in different load-bearing regions while there was no statistical difference for superficial cartilage. No significant change in T2* value of deep cartilage after activity was observed, while T2* values of the superficial layer decreased after activity in both the medial tibial and lateral femoral cartilage, indicating that different bio-mechanical properties exist within the deep and superficial cartilage.

4212		Intermediate exchange rate measured between bound and free water pools of tendon: implications on bi-component relaxation modeling
		Muhammad Ali Raza Anjum¹, Anshuman Swain ¹, Johannes Leisen², Felix Gonzalez¹, and David Reiter ¹
		¹Emory University, Atlanta, GA, United States, ²Georgia Tech, Atlanta, GA, United States
		Bi-component T2* modelling is widely applied to tendon to elucidate structural and biochemical properties of tissue. This model does not incorporate magnetization exchange between collagen-bound water and interstitial water pools. This study reports intermediate exchange-rates compared with respective pool relaxation rate estimates, in ex vivo bovine and ovine Achilles tendon, measured using T2–T2 correlation NMR. Intermediate exchange causes the bi-component model to underestimate the population of the rapidly-relaxing pool and overestimate relaxation rates and the population of the slowly-relaxing pool. The two-pool exchange model could be more specific to structural and biochemical changes in tendon with pathology.

4213		State of the ART (Adversarial Robust Training) to Reconstruct Clinically Relevant Features in Accelerated Knee MRI
		Francesco Caliva¹, Victor Kaiyang Cheng², Rutwik Shah¹, Misung Han¹, Sharmila Majumdar¹, and Valentina Pedoia¹
		¹University of California San Francisco, San Francisco, CA, United States, ²University of California Berkeley, Berkeley, CA, United States
		We propose an Adversarial Robust Training (ART) strategy to overcome the problem with accelerated MRI models, which are prone to missing small yet clinically relevant features. We introduced small, difficult to reconstruct synthetic features to undersampled MRIs and encouraged their reconstruction through robust training. To assess generalizability of our technique to real world applications, we annotated morphological features relevant to musculoskeletal disease diagnosis on images in the FastMRI dataset and tested ART. Overall, the approach has potential to reduce network instability and improve reliability and fidelity in image reconstruction.

4214		Quantitative Evaluation of Lumbar Bone Marrow Fat Content with Age，Gender and Body Mass Index by Using 3D mDixon Quant
		Yu Song¹, Qingwei qing Song², Yingkun Guo¹, Gang Ning¹, Xuesheng Li¹, and Yu Song³
		¹Department of Radiology, West China Second University Hospital, Sichuan University, Chengdu, China, ²Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian, China, ³Department of Radiology, West China Second University, Sichuan University, Chengdu, China
		The changes of adipose tissue in bone marrow will directly affect the changes of bone marrow composition. 3D mDixon Quant technology is a new method for precise fat quantification developed by Philips. It automatically generates high-quality FF and T2 * mapping maps by multi-echo acquisition combined with 7 peak fat model and T2 * correction, so as to achieve accurate and rapid quantification of bone marrow fat content.

4215		Dual-energy CT-based bone marrow imaging in multiple myeloma: Assessment of focal lesions in relation to disease status and MRI findings
		Sebastian Werner¹, Bernhard Krauss², and Marius Horger¹
		¹Radiology, University Hospital Tuebingen, Germany, Tuebingen, Germany, ²Siemens Healthineers, Forchheim, Germany
		Retrospective quantitative evaluation of 103 focal osteolytic lesions of the axial skeleton in virtual-non-calcium bone marrow images of 32 multiple myeloma patients in correlation to hematologic disease status, T1w signal intensity and ADC. Virtual-non-calcium bone marrow imaging allows differentiation between overall active and inactive disease with higher attenuation signifying an increasing likelihood of active disease. This is supported by a significant positive correlation between the attenuation and the ADC, as well as a corresponding inverse correlation to T1w signal intensity.

4216		Quantitative Magnetization Transfer MRI of the Knee Meniscus: Correlations with Biochemistry and Histology
		Kirstin D Olsen¹, Lumeng Cui¹, Brennan E Berryman², Ives R Levesque³, and Emily J McWalter²
		¹Biomedical Engineering, University of Saskatchewan, Saskatoon, SK, Canada, ²Mechanical Engineering, University of Saskatchewan, Saskatoon, SK, Canada, ³Medical Physics, McGill University, Montreal, QC, Canada
		The objective of this research was to determine if quantitative magnetization transfer (qMT) MRI can assess knee meniscus tissue health. Cadaver knee menisci underwent qMT assessment, biochemical content assessment, and histological scoring. Weak correlations were found between T_1f and collagen content, T_1f and histology score, and the histology scores and T_2f and T_2b. A moderate correlation was found between T_2fand collagen content. qMT shows promise as a measure of meniscal tissue health but future studies with more severely damaged tissue are required to elucidate relationships between qMT and measures of meniscal content and health.

4217		A multi-spectral imaging template of hip arthroplasty showing distribution of 3T vs. 1.5T artifacts in the acetabulum of varying implants
		Andrew S. Nencka¹, Peter S. Johnson¹, and Kevin M. Koch¹
		¹Radiology, Medical College of Wisconsin, Milwaukee, WI, United States
		An average proton density weighted template of the hip with arthroplasty was developed, and multi-spectral images from patients acquired at both 1.5T and 3.0T were registered to it. As a proof-of-concept, maps of residual metal artifact were transformed into the template space and, for different implant compositions, probability maps of residual artifact were generated. Ceramic implants exhibit minimal cross-field artifact differences, while metal-on-metal implants show more profound differences and metal-on-poly implants exhibit more moderate differences. This template and registration software lays the foundation for future quantitative analysis of cross-session and cross-subject imaging of hip arthroplasty.

4218		Fast T1 Measurement of Cortical Bone using 3D Ultrashort Echo Time Actual Flip Angle Imaging and Single Repetition Time Acquisition (UTE-AFI-STR)
		Zhao Wei^1,2,3, Hyungseok Jang¹, Zubiad Ibrahim¹, Mohammadamin Cheraghi¹, Graeme M. Bydder¹, Wenhui Yang^2,3, and Ya-Jun Ma¹
		¹Department of Radiology, UC San Diego, San Diego, CA, United States, ²Institute of Electrical Engineering, Chinese Academy of Sciences, Beijing, China, ³University of Chinese Academy of Sciences, Beijing, China
		To improve the time efficiency for cortical bone T₁ measurement on clinical scanners, we proposed a new method which fits the datasets of 3D ultrashort echo time actual flip angle imaging (UTE-AFI) and UTE with a single TR (UTE-STR) simultaneously (UTE-AFI-STR). The results show that the UTE-AFI-STR can measure the T₁ of cortical bone accurately and more efficiently than the UTE-AFI-VTR method.

4219		Quantitative 7T MRI for Post-traumatic Osteoarthritis Progression in a Rabbit Model
		Rossana Terracciano^1,2, Yareli Carcamo-Bahena¹, Xiaowei Zou³, Joshua D. Harris⁴, Bradley Weiner⁴, John Scott Labis⁵, Nakul Gupta⁵, and Carly S. Filgueira^1,6
		¹Nanomedicine, Houston Methodist Research Institute, Houston, TX, United States, ²Electronics and Telecommunications, Politecnico di Torino, Torino, Italy, ³Siemens Medical Solutions USA Inc, Malvern, PA, United States, ⁴Orthopedic Surgery, Houston Methodist Research Institute, Houston, TX, United States, ⁵Clinical Radiology, Houston Methodist Research Institute, Houston, TX, United States, ⁶Cardiovascular Surgery, Houston Methodist Research Institute, Houston, TX, United States
		No medical therapies have been shown to halt or slow the progression of post-traumatic osteoarthritis (PTOA), which develops after acute/repetitive injury to a joint. Improvements in non-invasive methods to assess PTOA will help diagnose stages for early clinical interventions. Rabbits offer a translationally relevant animal model where anterior cruciate ligament (ACL) injury allows for monitoring PTOA progression, determining a treatment window for early invention. In this work, we leveraged the high signal-to-noise ratio (SNR) of magnetic resonance imaging at 7T to develop a quantitative, non-invasive, 3D-ultrahigh-resolution dGEMRIC MRI exam for femoral-tibial observations and T1 quantification of matrix changes in rabbits.

4220		Quantitative T2 mapping MRI near metal: comparison of metal artifact reduction techniques
		Takayuki Sakai^1,2, Masami Yoneyama³, Atsuya Watanabe^4,5, Daichi Murayama¹, Shigehiro Ochi¹, and Tosiaki Miyati⁶
		¹Radiology, Eastern Chiba Medical Center, Tonage, Japan, ²Division of Health Sciences, Graduate School of Medical Sciences, Kanazawa University, Kanazawa, Japan, ³Philips Japan, Tokyo, Japan, ⁴General Medical Services, Chiba University Graduate School of Medicine, Chiba, Japan, ⁵Orthopaedic Surgery, Eastern Chiba Medical Center, Chiba, Japan, ⁶Faculty of Health Sciences, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University, Kanazawa, Japan
		Cartilage degeneration can be evaluated non-invasively by qualitative T2 mapping MRI. Although cartilage evaluation using T2 mapping is desirable in cases which open reduction and internal fixation for fractures around the knee joint and osteotomy for knee osteoarthritis, evaluation is often difficult due to the existence of metal artifacts. Theoretically, metal artifacts are suppressed by metal artifact reduction techniques such as VAT and SEMAC. Therefore, we evaluated the usefulness of T2 mapping using VAT or SEMAC for cartilage around the knee joint that was metal-fixed after surgery.

4221		Monitoring the efficacy of TNF-α antagonists in the treatment of SpA preliminary study based on MRI biochemical Imaging technique
		Yu Shun¹, Lin Min-gui¹, Zhongshuai Zhang², Chen Xian-yuan¹, and Ma Mingping¹
		¹Radiology department, Fujian Provincial Hospital, fuzhou, China, ²Diagnostic Imaging, SIEMENS Healthcare, Shanghai, China
		This study aimed to evaluate the feasibility of using MRI to monitor the efficacy of tumor necrosis factor-α antagonists in the treatment of SpA (axial spondyloarthritis).

Quantitative MRI II

Quantitative MSK MRI

Musculoskeletal

4222		Vertebral bone marrow water T2 is sex-dependent and negatively correlated with age and the proton density fat fraction (PDFF)
		Stefan Ruschke¹, Jan Syväri¹, Michael Dieckmeyer², Daniela Junker¹, Marcus R. Makowski¹, Thomas Baum², and Dimitrios C. Karampinos¹
		¹Department of Diagnostic and Interventional Radiology, School of Medicine, Technical University of Munich, Munich, Germany, ²Department of Diagnostic and Interventional Neuroradiology, School of Medicine, Technical University of Munich, Munich, Germany
		Single-voxel multi-TE STEAM MRS data of the 4^th/5^th lumber vertebrae from 260 subjects (age: 0.7-77.7years) was retrospectively analyzed to investigate physiological variations of water T2 (T2W) in bone marrow. The data was fitted using a joint-series T2-constrained time domain-based water–fat signal model. Linear regression analysis indicated significant negative correlations for T2W vs. age and PDFF for both sexes, respectively. Females showed significant longer T2W values compared to males. Multiple linear regression revealed PDFF and sex as significant predictors of T2W. Underlying physiological variations of T2W are of relevance in the study of T2 variations and T2-weighted parameters.

4223		Quantitative Assessment of cartilage composition using MR T1ρ and T2 imaging 10 years Post ACL-Reconstruction
		John P. Murray^1,2,3, Richard Lartey^1,3, Jeehun Kim^1,3, Mei Li^1,3, Sibaji Gaj^1,3, Brendan Eck^1,4, Donxing Xie^1,3, Carl Winalski^1,3,4, Faysal Altahawi^3,4, Morgan H. Jones^3,5, Bruce M Damon⁶, Laura J Huston⁷, Huyen T. Nguyen⁸, Michael V Knopp⁸, Kurt P. Spindler^3,5, and Xiaojuan Li^1,3,4
		¹Dept. of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, Cleveland, OH, United States, ²Dept. of Biomedical Engineering, Case Western Reserve University, Cleveland, OH, United States, ³Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, OH, United States, ⁴Dept. of Diagnostic Radiology, Imaging Inst., Cleveland Clinic, Cleveland, OH, United States, ⁵Dept. of Orthopaedic Surgery, Orthopaedics and Rheumatology Inst., Cleveland, OH, United States, ⁶Dept. of Radiology, Vanderbilt University, Nashville, TN, United States, ⁷Dept. of Orthopaedics and Rehabilitation, Vanderbilt University, Nashville, TN, United States, ⁸Dept. of Radiology, Wright Ctr. Of Innovation in BioMed. Imaging, The Ohio State University, Columbus, OH, United States
		This is a preliminary report for the MOON multivendor multisite 10 year follow up using quantitative MR imaging techniques for patients following ACL reconstruction. Imaging protocol was harmonized between sites and T1ρ and T2 imaging sequences that were cross-validated between sites and vendors were applied. Phantom data showed excellent intra-site repeatability and small inter-site variations. Significantly elevated cartilage T1ρ and T2 were observed in operated knees compared to contralateral knees. The preliminary results demonstrated feasibility of combining multi-site large cohort study with advanced quantitative MRI, with harmonized protocols and rigorous quality control for both data acquisition and processing.

4224		How radiology and clinical severity of knee osteoarthritis correlate: Analysis with cartilage T1rho and T2 values, WORMS, and K-L grade
		Woo Young Kang¹, Suk-Joo Hong¹, Jinwoo Han¹, Yoonmi Choi¹, Chang Ho Kang², Kyung-sik Ahn², Baek Hyun Kim³, and Euddeum Shim³
		¹Radiology, Korea University Guro Hospital, Seoul, Korea, Republic of, ²Radiology, Korea University Anam Hospital, Seoul, Korea, Republic of, ³Radiology, Korea University Ansan Hospital, Ansan, Korea, Republic of
		Recently, MRI is increasingly used as an imaging tool to evaluate knee osteoarthritis. The purpose of this study is not only to evaluate the structural and biomechanical changes in osteoarthritis patients on MRI, but also to compare and analyze the imaging features and quantitative measurements evaluated by MRI with clinical severity.

4225		The diagnostic value of a new indicator in the chronic anterior cruciate ligament injury based on Ultra-Short Echo imaging
		Yanjun Hu¹, Cheng Xu¹, Yexin He¹, Ruizhu Wang¹, and Kaiyu Wang²
		¹Magnetic Resonance Center, Shanxi Provincial People’s Hospital, Taiyuan, China, ²MR Research, GE Healthcare, Beijing, China
		At present, whether patients with anterior cruciate ligament (ACL) injury can return to exercise lacks quantitative imaging indicators. In this study, ultra-short echo technique was used to evaluate the change of short T2 components like ligament in patients at chronic stage of ACL injury, finding that the proportion of short T2 componets (POS) in the overall signal can be used to reflect the extent of ACL injury and recovery. This study suggests that there are changes in the short-T2 components in patients with chronic ACL injury, and POS is superior to IKDC in assessing joint stability.

4226		Repeatability and orientation dependence of ultrashort echo time (UTE) T2* mapping at 3T for the whole knee
		Zhenzhou Wu¹, Stefan Sommer^2,3, Xiaodong Zhong⁴, Kecheng Liu⁴, Jeehun Kim¹, Jillian Beveridge¹, Xiaoliang Zhang⁵, and Xiaojuan Li¹
		¹Program of Advanced Musculoskeletal Imaging (PAMI), Cleveland Clinic, Cleveland, OH, United States, ²Siemens Healthcare, Zurich, Switzerland, ³Swiss Center for Musculoskeletal Imaging (SCMI), Balgrist Campus, Zurich, Switzerland, ⁴Siemens Medical Solutions USA, Inc., Malvern, PA, United States, ⁵Department of Biomedical Engineering, University at Buffalo, State University of New York, Buffalo, NY, United States
		The connective tissues in the knee joint have ultrashort MR T2* relaxation times. Ultrashort echo time (UTE) sequences can offer a unique tool for measuring the fast-decaying signals in these tissues. However, T2* measures in these collagen-rich tissues are subject to magic angle effect. Previous studies on such orientation dependence and repeatability of UTE T2* in the knee are limited. The objectives of this study were to evaluate the repeatability of UTE T2, and to investigate and compare the orientation dependence of T2 mapping between UTE and regular gradient echo (GRE) imaging sequences for whole knee imaging.

4227		Three dimensional adiabatic T1ρ prepared ultrashort echo time Cones (3D UTE-Cones-AdiabT1ρ) imaging of knee joint degeneration
		Mei Wu^1,2, Yajun Ma¹, Guanyuan Ning², Saeed Jerban¹, Yanping Xue¹, Zhao Wei¹, Eric Y Chang^1,3, and Jiang Du¹
		¹Department of Radiology, University of California San Diego, San Diego, CA, United States, ²Department of Radiology, Guangzhou First People’s Hospital, School of Medicine, South China University of Technology, Guangzhou, China, ³Radiology Service, VA San Diego Healthcare System, San Diego, CA, United States
		The three-dimensional adiabatic T1ρ prepared ultrashort echo time Cones (3D-UTE-Cones-AdiabT1ρ) sequence is a novel imaging technique that can provide magic angle-insensitive assessment of proteoglycan depletion in both short and long T2 tissues in the knee joint. We applied this sequence to healthy volunteers and patients with different degrees of OA for a systematic evaluation of its clinical performance. Results showed that the sequence could be used for quantitative evaluation of the knee cartilage degeneration, and that the 3D-UTE-Cones-AdiabT1ρ showed a significant positive relationship with WORMS and KL score，and significant difference in different extent and depth lesions of cartilage.

4228		Evaluation of Dixon MRI Methods for Quantitative Assessment of Thigh Muscle Fatty Infiltration in Post-Traumatic Osteoarthritis
		Brendan L. Eck^1,2, Richard Lartey^1,3, Dongxing Xie^1,3, Jeehun Kim^1,3, Carl S. Winalski^1,2,3, Bruce M. Damon⁴, Xiaodong Zhong⁵, Kecheng Liu⁵, Dimitris Karampinos⁶, Faysal Altahawi^1,2, Morgan H. Jones^1,7, Kurt P. Spindler^1,7, and Xiaojuan Li^1,2,3
		¹Program of Advanced Musculoskeletal Imaging, Cleveland Clinic, Cleveland, OH, United States, ²Diagnostic Radiology, Cleveland Clinic, Cleveland, OH, United States, ³Biomedical Engineering, Cleveland Clinic, Cleveland, OH, United States, ⁴Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, TN, United States, ⁵MR R&D Collaborations, Siemens Medical Solutions USA, Inc., Malvern, PA, United States, ⁶Department of Diagnostic and Interventional Radiology, School of Medicine, Technical University of Munich, Munich, Germany, ⁷Orthopaedic Surgery, Orthopaedics and Rheumatology Institute, Cleveland Clinic, Cleveland, OH, United States
		Fatty infiltration in thigh skeletal muscle is a potential biomarker of osteoarthritis and post-traumatic osteoarthritis. Quantification of fatty infiltration is possible by Dixon MRI but is dependent on acquisition and processing. We evaluated five acquisition and processing techniques for reproducibility in phantom and healthy controls. Patients at 10-years post-ACL reconstruction were scanned to evaluate the potential for these techniques to detect thigh muscle fatty infiltration. Monopolar gradient acquisition and magnitude image-based processing improved the robustness of fat fraction quantification. Vendor-independent magnitude-based processing and vendor inline processing similarly quantified elevated fat fraction in the hamstring muscles of patients' ACL reconstructed legs.

4229		Improvement of distortion-free diffusion tensor imaging of lumbar nerve roots using direct coronal MultiVane-SPLICE diffusion-weighted MRI.
		Takayuki Sakai^1,2, Masami Yoneyama³, Atsuya Watanabe^4,5, Daichi Murayama¹, Shigehiro Ochi¹, and Tosiaki Miyati⁶
		¹Radiology, Eastern Chiba Medical Center, Tonage, Japan, ²Division of Health Sciences, Graduate School of Medical Sciences, Kanazawa University, Kanazawa, Japan, ³Philips Japan, Tokyo, Japan, ⁴General Medical Services, Chiba University Graduate School of Medicine, Chiba, Japan, ⁵Orthopaedic Surgery, Eastern Chiba Medical Center, Chiba, Japan, ⁶Faculty of Health Sciences, Institute of Medical, Pharmaceutical and Health Sciences, Kanazawa University, Kanazawa, Japan
		Diffusion tensor imaging (DTI) has a severe problem such as high geometric distortion and body motion. To solve these problems, we attempted to apply a split acquisition of fast spin-echo signals for diffusion imaging (SPLICE) instead of conventional Mx-eliminated TSE-DTI. In addition, so as to improve the incoherent signal phase due to motion, SPLICE was combined with MultiVane (known as PROPELLER) which is to sample k-space in a rotating fashion using a set of radially directed blades. In this study, we evaluated the effect of accuracy and reproducibility of the quantitative values in lumbar nerve roots using MultiVane-SPLICE-DTI.

4230		Can acquisition of pretreatment IVIM parameters in tumor predict the outcomes of osteosarcoma treated by chemotherapy? A preliminary study
		Esha Baidya Kayal¹, Devasenathipathy Kandasamy², Kedar Khare³, Raju Sharma², Sameer Bakhshi⁴, and Amit Mehndiratta^1,5
		¹Centre for Biomedical Engineering, Indian Institute of Technology Delhi, New Delhi, India, ²Department of Radio diagnosis, All India Institute of Medical Sciences Delhi, New Delhi, India, ³Department of Physics, Indian Institute of Technology Delhi, New Delhi, India, ⁴Department of Medical Oncology, Dr. B.R. Ambedkar Institute-Rotary Cancer Hospital (IRCH), All India Institute of Medical Sciences Delhi, New Delhi, India, ⁵Department of Biomedical Engineering, All India Institute of Medical Sciences Delhi, New Delhi, India
		The aim of this study was to investigate the potential of Intravoxel incoherent motion (IVIM) parameters to predict event-free-survival (EFS) in patients with localized osteosarcoma treated by chemotherapy. Quantitative and IVIM parameter estimation and histogram analysis was performed in tumor volume at baseline. Imaging parameters along with clinical features were analyzed to assess their effects on EFS. Results showed, 25^th-percentile of perfusion coefficient (D*) and serum albumin level were significantly associated with EFS among localized osteosarcoma patients.

4231		Evaluation of Lumbar Disc Herniation Using Ultrashort Echo Time Magnetization Transfer in Posterior Longitudinal Ligament and Nucleus Pulposus
		Jianwei Liao¹, Jin Liu¹, Yajun Ma², Xiaojun Chen¹, Wei Li¹, Lin Yao¹, Long Qian³, Jiang Du², and Shaolin Li¹
		¹Department of Radiology, The Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, China, ²Department of Radiology, University of California, San Diego, CA, United States, ³MR Research, GE Healthcare, Guangzhou, China
		Magnetization transfer ratio (MTR) has been used for assessment of macromolecules in biological tissues. Ultrashort echo time technique magnetization transfer (UTE-MT) can quantify both of short and long T2 tissues, may have potential to evaluate early disc herniation. The current study aims to assess the feasibility of UTE-MT quantification of posterior longitudinal ligament and nucleus pulposus for evaluation of early disc herniation. It was concluded that the UTE-MT sequence is feasible to quantify posterior longitudinal ligament and nucleus pulposus in lumbar. Posterior longitudinal ligament and nucleus pulposus UTE-MTR may provide a new opportunity for early prediction of lumbar disc herniation.

4232		Imaging Sign Of Edema As A Possible Diagnostic Aid For Lateral Lumbar Spinal Canal Stenosis
		Shi Yin¹, Dou Weiqiang², and Ding Hongyuan¹
		¹The First Affiliated Hospital of Nanjing Medical University, Nanjing, China, ²GE Healthcare, MR Research, Beijing, P.R. China, Beijing, China
		This study aims to if quantitative MR DTI technology combined with conventional MR can be applied to increase diagnostic efficiency and improve the diagnostic accuracy for lateral lumbar spinal canal stenosis in clinical. By measuring 99 patients, we found in extraspinal sub-regions of Lumbar nerve root, imaging findings of nerve edema, increased diameter and decreased FA values have an ideal diagnostic performance in LLSCS. Besides, among all include parameters, nerve edema demonstrated the best diagnostic performance. Therefore, conventional MR combined with quantitative MR DTI/DTT has a good performance for visualizing and quantitative diagnosis in lateral lumbar spinal canal stenosis.

4233		Assessment of Osteoporosis in Human Lumbar Using 3D Adiabatic Inversion Recovery Prepared Ultrashort TE Cones Sequence
		Jin Liu¹, Yajun Ma², Jianwei Liao¹, Xiaojun Chen¹, Wei Li¹, Lin Yao¹, Long Qian³, Jiang Du², and Shaolin Li¹
		¹Department of Radiology, The Fifth Affiliated Hospital of Sun Yat-Sen University, Zhuhai, China, ²Department of Radiology, University of California, San Diego, CA, United States, ³MR Research, GE Healthcare, Guangzhou, China
		The 3D adiabatic inversion recovery prepared ultrashort TE cones (3D IR-UTE) sequence can image and quantify proton density assessment of short T2 water components in trabecular bone in vivo, and has the potential to diagnosis people with osteoporosis. The current study aims to prospective investigate the performance in diagnosing osteoporosis of IR-UTE sequence in human lumbar and to compare with quantitative computed tomography (QCT), dual-energy X-ray absorptiometry (DXA) and Fracture Risk Assessment Tool (FRAX) scores. It was concluded that IR-UTE was positively correlated to QCT and DXA, and was negatively correlated to FRAX scores.

4234		B1 Field Inhomogeneity Correction for qDESS T2 Mapping: Application to Rapid Bilateral Knee Imaging
		Marco Barbieri¹, Lauren Watkins^1,2, Arjun D. Desai^1,3, Valentina Mazzoli¹, Elka Rubin¹, Andrew Schmidt¹, Garry E. Gold^1,2, Brian A. Hargreaves^1,2,3, Akshay S. Chaudhari^1,4, and Feliks Kogan¹
		¹Department of Radiology, Stanford University, Stanford, CA, United States, ²Department of Bioengineering, Stanford University, Stanford, CA, United States, ³Department of Electrical Engineering, Stanford University, Stanford, CA, United States, ⁴Department of Biomedical Data Science, Stanford University, Stanford, CA, United States
		Bilateral T₂ mapping is a powerful tool for studying OA changes and investigating the role of between-knee asymmetry. qDESS allows for rapid T₂ mapping but uses a model that requires knowledge of the FA. Hence, B₁ inhomogeneities may affect the accuracy of T₂ measurements. We propose a pixel-wise B₁-correction method for qDESS T₂ mapping exploiting an auxiliary B₁ map to compute the actual FA used in the model. The technique was validated with phantom and with T₂ measurements of femoral cartilage in simultaneous bilateral knee imaging. The results showed that B1-correction can mitigate T₂ variations that were driven by B1-inhomogeneities.

4235		Denoising Meniscus T2* Mapping In College Basketball Players
		Ek T Tan¹, Erin C Argentieri¹, Madeleine A Gao¹, John Neri¹, Garry E Gold², Sharmila Majumdar³, Hollis G Potter¹, and Matthew F Koff¹
		¹Radiology and Imaging, Hospital for Special Surgery, New York, NY, United States, ²Stanford University, Stanford, CA, United States, ³University of California San Francisco, San Francisco, CA, United States
		Ultra-short echo time (UTE) T2* relaxometry is sensitive to meniscal injury, but suffers from low SNR when acquired at high spatial resolution. Two denoising techniques – principal components analysis (PCA) denoising and model-based denoising (MD) were evaluated. Simulations show PCA and MD work complimentarily to reduce variance and bias. In N=14 college basketball (cases) players vs. N=13 controls, MD and MDPCA significantly reduced coefficient of variance in the lateral menisci. A significant difference in T2* of lateral and menisci was observed in cases vs. controls.

4236		Changes in meniscus T2 relaxation times due to acute exercise in individuals with knee osteoarthritis
		Ananya Goyal¹, Garry Gold¹, Feliks Kogan¹, and Lauren Watkins¹
		¹Stanford University, Stanford, CA, United States
		Osteoarthritis (OA) is a debilitating disease characterized by joint structural degradation. Meniscal breakdown is linked to OA progression and can be studied through T2 relaxometry. In this work, we assess the effect of acute exercise on meniscus T2 relaxation times in individuals with OA. 12 individuals were scanned before and after performing a single-legged exercise. T2 relaxation times were computed using a qDESS sequence. We found that acute exercise did not affect the meniscal T2 values in OA subjects; however, significant changes in the maximal regional T2 change after exercise were observed that may be impacted by knee loading.

4237		Kinematic MRI Tracking of Wrist Carpal Bones
		Mohammad Zarenia¹, Volkan Emre Arpinar¹, Andrew S. Nencka¹, L. Tugan Muftuler², and Kevin M. Koch¹
		¹Department of Radiology, Medical College of Wisconsin, Milwaukee, WI, United States, ²Department of Neurosurgery, Medical College of Wisconsin, Milwaukee, WI, United States
		In this study, a methodology is developed for kinematic tracking and profiling of free-moving wrist carpal bones. The derived kinematic profiles consist of joint metrics computed and tracked during free wrist motion. Such metrics are hypothesized to be potentially valuable for diagnosis and management of joint dysfunction. Here, an advanced 4D MRI protocol is deployed to capture frames of unconstrained moving joints which are registered to a high-resolution fixed volume using a novel slab-to-volume boundary registration method. Using this registration approach on a sample subject-dataset, concept demonstrations of a selection of metrics describing scaphoid-lunate kinematic motion are presented.

4238		Whole-body MRI reveals the burden of unsuspected synovitis in juvenile idiopathic arthritis
		Varvara Choida^1,2, Timothy J.P. Bray^1,3, Alan Bainbridge⁴, Debajit Sen^2,5, Corinne Fisher^2,5, Maria Leandro^2,5, Coziana Ciurtin^2,5, and Margaret Hall-Craggs^1,3
		¹Centre for Medical Imaging, University College London, London, United Kingdom, ²Centre for Adolescent Rheumatology Versus Arthritis at UCL UCLH and GOSH, University College London, London, United Kingdom, ³Radiology, University College London Hospital, London, United Kingdom, ⁴Medical Physics, University College London Hospitals, London, United Kingdom, ⁵Adolescent and young adult Rheumatology, University College London Hospital, London, United Kingdom
		MRI is used for the detection of synovitis in juvenile idiopathic arthritis (JIA). We developed a whole-body MRI protocol, consisting of coronal, post-contrast mDixon images, which is fast and offers good coverage of peripheral joints and spine. A prospective study including 32 adolescent and young adult patients with JIA revealed that 43.8% of patients had synovitis in at least one joint on imaging that was not detectable clinically. The proportion of patients with subclinical synovitis was similar in clinically active and inactive patients. This could lead to identification of unsuspected inflammation in JIA patients with potential impact on their treatment.

4239		Gadolinium-Free Assessment of Synovitis Using Diffusion Tensor Imaging
		Halston J.C. Sandford¹, James W. MacKay², Lauren E. Watkins^1,3, Garry E. Gold^1,3,4, Feliks Kogan¹, and Valentina Mazzoli¹
		¹Radiology, Stanford University, Stanford, CA, United States, ²University of East Anglia, Norwich, United Kingdom, ³Bioengineering, Stanford University, Stanford, CA, United States, ⁴Orthopaedic Surgery, Stanford University, Stanford, CA, United States
		Synovitis is associated with osteophyte development and osteoarthritis progression. The intensity of synovitis can be quantified by K^trans, obtained from DCE-MRI using gadolinium-based contrast agent. Diffusion Tensor Imaging (DTI) has shown potential for detecting synovitis without administering contrast agent. This study assessed the potential of DTI to quantify synovitis by comparing DTI parameters (MD, FA) to K^trans within the whole synovium and adjacent to osteophytes. We found that FA negatively associates with K^trans in both regions, while MD has a positive correlation only within the whole synovium. DTI is a promising tool for assessing intensity of synovitis in osteoarthritic knees.

4240		Optimizing the use of diffusion tensor imaging for clinical tractography of the anterior cruciate ligament in the knee
		Allen A Champagne^1,2, Don Brien², Andrew McGuire³, Paul Fenton⁴, Yousef A Marwan⁵, Paul A Martineau⁵, and Davide D Bardana³
		¹School of Medicine, Queen's University, Kingston, ON, Canada, ²Center for Neuroscience Studies, Queen's University, Kingston, ON, Canada, ³Orthopedic Surgery, Queen's University, Kingston, ON, Canada, ⁴Diagnostic Radiology, Queen's University, Kingston, ON, Canada, ⁵Orthopedic Surgery, McGill University, Montreal, QC, Canada
		Emergence of diffusion tensor imaging (DTI) has provided an exciting avenue to characterize the ligamentization process of the anterior cruciate ligament (ACL), following knee injury. However, a diverse, but limited, suite of published diffusion weighted sequences with varying parameters have posed a significant challenge to clinician-scientists wanting to integrate DTI as a tool to study the ACL. In this study, we provide a detailed comparison of DTI acquisitions toward the optimisation of ACL reconstruction using semi-automated fiber-based probabilistic tractography (FBPT), as a step-stone and reference for the clinical integration of DTI toward assessing microstructural integrity of the cruciate ligaments.

4241		Assessment of Synovitis in Osteoarthritis Using Non-Contrast Quantitative DESS
		Jacob Thoenen¹, James W. MacKay^2,3, Kathryn J. Stevens¹, Tom D. Turmezei⁴, Akshay Chaudhari¹, Lauren E. Watkins¹, Emily J. McWalter⁵, Brian A. Hargreaves¹, Garry E. Gold¹, and Feliks Kogan¹
		¹Department of Radiology, Stanford University, Stanford, CA, United States, ²Department of Radiology, University of Cambridge, Cambridge, United Kingdom, ³Norwich Medical School, University of East Anglia, Norwich, United Kingdom, ⁴Department of Radiology, Norfolk and Norwich University Hospital, Norwich, United Kingdom, ⁵Department of Mechanical Engineering, University of Saskatchewan, Saskatoon, SK, Canada
		Quantitative double-echo in steady-state (qDESS) MRI has been proposed as an alternative to T1-weighted contrast enhanced MRI (CE-MRI) for evaluation of synovitis. In comparison with CE-MRI by, four radiologists, qDESS with a low diffusion gradient (qDESS_Low), and qDESS with a high diffusion gradient (qDESS_High) both showed good agreement to CE-MRI for the overall and regional impressions of synovitis, as well as good inter- and intra- rater agreement. Further, radiologists rated qDESS_Low and qDESS_High with moderate-to-high and moderate diagnostic confidence, respectively. This work shows the potential of the qDESS sequence as a non-contrast imaging technique for evaluation of synovitis.