Challenges in MRI Liver Interpretation: Emerging Solutions
Koichiro Yasaka1
1Radiology, The Institute of Medical Science, The University of Tokyo, Tokyo, Japan

Synopsis

This talk will introduce the emerging solutions for MRI liver interpretations in liver fibrosis staging and liver mass differentiation, especially those based on radiomics strategies and deep learning technique.

Overview

This talk will introduce the emerging solutions for MRI liver interpretations in liver fibrosis staging and liver mass differentiation, especially those based on radiomics strategies [1] and deep learning technique [2-4].

Liver cirrhosis is an end stage of liver fibrosis, for which chronic infection by hepatitis viruses can be a risk factor. Liver cirrhosis is associated with deformation of liver morphology. By subjectively evaluating this on MRI images, by using expanded gallbladder fossa sign [5] etc., liver cirrhosis could be diagnosed. However, it has been difficult to diagnose early stages of liver fibrosis by such subjective interpretation. Recently, radiomics strategies, which utilize quantitatively evaluated texture parameters, have been gaining much attention [1]. By incorporating quantitative texture parameters into a logistic regression model, early stages of liver fibrosis were enabled to be diagnosed [6]. Most recently, deep learning approaches are also reportedly effective in diagnosing F2–4 of liver fibrosis [7].

Several types of tumors arise in the liver. Dynamic CT and MRI are known to be effective in detailed evaluations of them. While subjective interpretation of liver mass MRI images is affected by radiologists’ experience, deep learning algorithms are reportedly able to semi-automatically differentiate them [8, 9] or categorize them according to LI-RADS [10]. As for the dynamic study of MRI examinations of the liver, hepatocyte-specific contrast materials are used. However, use of this contrast material was known to be associated with motion artifact in arterial phase [11] due to rapid tachypnea and breath-hold failure [12, 13]. Deep learning algorithm is known to be effective in reducing such artifacts [14].

Acknowledgements

None.

References

1. Gillies RJ, Kinahan PE, Hricak H. Radiomics: Images Are More than Pictures, They Are Data. Radiology. 2016;278(2):563-77.

2. Yasaka K, Akai H, Kunimatsu A, Kiryu S, Abe O. Deep learning with convolutional neural network in radiology. Jpn J Radiol. 2018;36(4):257-272.

3. Yasaka K, Abe O. Deep learning and artificial intelligence in radiology: Current applications and future directions. PLoS Med. 2018;15(11):e1002707.

4. Chartrand G, Cheng PM, Vorontsov E, Drozdzal M, Turcotte S, Pal CJ, Kadoury S, Tang A. Deep Learning: A Primer for Radiologists. Radiographics. 2017;37(7):2113-2131.

5. Ito K, Mitchell DG, Gabata T, Hussain SM. Expanded gallbladder fossa: simple MR imaging sign of cirrhosis. Radiology. 1999;211(3):723-6.

6. Park HJ, Lee SS, Park B, Yun J, Sung YS, Shim WH, Shin YM, Kim SY, Lee SJ, Lee MG. Radiomics Analysis of Gadoxetic Acid-enhanced MRI for Staging Liver Fibrosis. Radiology. 2019 Feb;290(2):380-387.

7. Yasaka K, Akai H, Kunimatsu A, Abe O, Kiryu S. Liver Fibrosis: Deep Convolutional Neural Network for Staging by Using Gadoxetic Acid-enhanced Hepatobiliary Phase MR Images. Radiology. 2018;287(1):146-155.

8. Yasaka K, Akai H, Abe O, Kiryu S. Deep Learning with Convolutional Neural Network for Differentiation of Liver Masses at Dynamic Contrast-enhanced CT: A Preliminary Study. Radiology. 2018;286(3):887-896.

9. Hamm CA, Wang CJ, Savic LJ, Ferrante M, Schobert I, Schlachter T, Lin M, Duncan JS, Weinreb JC, Chapiro J, Letzen B. Deep learning for liver tumor diagnosis part I: development of a convolutional neural network classifier for multi-phasic MRI. Eur Radiol. 2019;29(7):3338-3347.

10. Yamashita R, Mittendorf A, Zhu Z, Fowler KJ, Santillan CS, Sirlin CB, Bashir MR, Do RKG. Deep convolutional neural network applied to the liver imaging reporting and data system (LI-RADS) version 2014 category classification: a pilot study. Abdom Radiol (NY). 2019. doi: 10.1007/s00261-019-02306-7. [Epub ahead of print]

11. Davenport MS, Viglianti BL, Al-Hawary MM, Caoili EM, Kaza RK, Liu PS, Maturen KE, Chenevert TL, Hussain HK. Comparison of acute transient dyspnea after intravenous administration of gadoxetate disodium and gadobenate dimeglumine: effect on arterial phase image quality. Radiology. 2013;266(2):452-61.

12. Akai H, Yasaka K, Nojima M, Kunimatsu A, Inoue Y, Abe O, Ohtomo K, Kiryu S. Gadoxetate disodium-induced tachypnoea and the effect of dilution method: a proof-of-concept study in mice. Eur Radiol. 2018;28(2):692-697.

13. Motosugi U, Bannas P, Bookwalter CA, Sano K, Reeder SB. An Investigation of Transient Severe Motion Related to Gadoxetic Acid-enhanced MR Imaging. Radiology. 2016;279(1):93-102.

14. Tamada D, Kromrey ML, Ichikawa S, Onishi H, Motosugi U. Motion Artifact Reduction Using a Convolutional Neural Network for Dynamic Contrast Enhanced MR Imaging of the Liver. Magn Reson Med Sci. 2019. doi: 10.2463/mrms.mp.2018-0156. [Epub ahead of print]

Proc. Intl. Soc. Mag. Reson. Med. 28 (2020)