The summaries are free for public
use. The Chronic Liver Disease
Foundation will continue to add and
archive summaries of articles deemed
relevant to CLDF by the Board of
Trustees and its Advisors.
Abstract Details
Deep learning-assisted differentiation of pathologically proven atypical and typical hepatocellular carcinoma (HCC) versus non-HCC on contrast-enhanced MRI of the liver
Eur Radiol. 2021 Jan 6. doi: 10.1007/s00330-020-07559-1. Online ahead of print.
Paula M Oestmann123, Clinton J Wang14, Lynn J Savic12, Charlie A Hamm12, Sophie Stark125, Isabel Schobert12, Bernhard Gebauer2, Todd Schlachter1, MingDe Lin1, Jeffrey C Weinreb1, Ramesh Batra6, David Mulligan6, Xuchen Zhang7, James S Duncan14, Julius Chapiro8
Author information
1Department of Radiology and Biomedical Imaging, Yale School of Medicine, 333 Cedar Street, New Haven, CT, 06520, USA.
2Institute of Radiology, Berlin Institute of Health, Charité - Universitätsmedizin Berlin, corporate member of Freie Universität Berlin, Humboldt-Universität, 10117, Berlin, Germany.
3Faculty of Medicine, Heinrich-Heine-University Düsseldorf, Düsseldorf, Germany.
4Department of Biomedical Engineering, Yale School of Engineering and Applied Science, New Haven, CT, 06520, USA.
5Faculty of Medicine, Albert-Ludwigs-University Freiburg, Freiburg, Germany.
6Department of Transplantation and Immunology, 333 Cedar Street, New Haven, CT, 06520, USA.
7Department of Pathology, Yale School of Medicine, 310 Cedar Street, New Haven, CT, 06520, USA.
8Department of Radiology and Biomedical Imaging, Yale School of Medicine, 333 Cedar Street, New Haven, CT, 06520, USA. julius.chapiro@yale.edu.
Abstract
Objectives: To train a deep learning model to differentiate between pathologically proven hepatocellular carcinoma (HCC) and non-HCC lesions including lesions with atypical imaging features on MRI.
Methods: This IRB-approved retrospective study included 118 patients with 150 lesions (93 (62%) HCC and 57 (38%) non-HCC) pathologically confirmed through biopsies (n = 72), resections (n = 29), liver transplants (n = 46), and autopsies (n = 3). Forty-seven percent of HCC lesions showed atypical imaging features (not meeting Liver Imaging Reporting and Data System [LI-RADS] criteria for definitive HCC/LR5). A 3D convolutional neural network (CNN) was trained on 140 lesions and tested for its ability to classify the 10 remaining lesions (5 HCC/5 non-HCC). Performance of the model was averaged over 150 runs with random sub-sampling to provide class-balanced test sets. A lesion grading system was developed to demonstrate the similarity between atypical HCC and non-HCC lesions prone to misclassification by the CNN.
Results: The CNN demonstrated an overall accuracy of 87.3%. Sensitivities/specificities for HCC and non-HCC lesions were 92.7%/82.0% and 82.0%/92.7%, respectively. The area under the receiver operating curve was 0.912. CNN's performance was correlated with the lesion grading system, becoming less accurate the more atypical imaging features the lesions showed.
Conclusion: This study provides proof-of-concept for CNN-based classification of both typical- and atypical-appearing HCC lesions on multi-phasic MRI, utilizing pathologically confirmed lesions as "ground truth."
Key points: • A CNN trained on atypical appearing pathologically proven HCC lesions not meeting LI-RADS criteria for definitive HCC (LR5) can correctly differentiate HCC lesions from other liver malignancies, potentially expanding the role of image-based diagnosis in primary liver cancer with atypical features. • The trained CNN demonstrated an overall accuracy of 87.3% and a computational time of < 3 ms which paves the way for clinical application as a decision support instrument.