Automated MRI liver segmentation for anatomical segmentation, liver volumetry, and the extraction of radiomics

作者：Moritz Gross, Steffen Huber, Sandeep Arora, Tal Zeevi, Stefan P. Haider, Ahmet S. Kücükkaya, Simon Iseke, Tom N. Kuhn, Bernhard Gebauer, Florian Michallek, Marc Dewey, Valérie Vilgrain, Riccardo Sartoris, Maxime Ronot, Ariel Jaffe, Mario Strazzabosco, Julius Chapiro, John A. Onofrey · 发表于：European Radiology · 年份：2024 · DOI：10.1007/s00330-023-10495-5 · 被引用次数：38 · 研究领域：Hepatocellular Carcinoma Treatment and Prognosis、Radiomics and Machine Learning in Medical Imaging、Liver Disease Diagnosis and Treatment

OBJECTIVES: To develop and evaluate a deep convolutional neural network (DCNN) for automated liver segmentation, volumetry, and radiomic feature extraction on contrast-enhanced portal venous phase magnetic resonance imaging (MRI). MATERIALS AND METHODS: This retrospective study included hepatocellular carcinoma patients from an institutional database with portal venous MRI. After manual segmentation, the data was randomly split into independent training, validation, and internal testing sets. From a collaborating institution, de-identified scans were used for external testing. The public LiverHccSeg dataset was used for further external validation. A 3D DCNN was trained to automatically segment the liver. Segmentation accuracy was quantified by the Dice similarity coefficient (DSC) with respect to manual segmentation. A Mann-Whitney U test was used to compare the internal and external test sets. Agreement of volumetry and radiomic features was assessed using the intraclass correlation coefficient (ICC). RESULTS: In total, 470 patients met the inclusion criteria (63.9±8.2 years; 376 males) and 20 patients were used for external validation (41±12 years; 13 males). DSC segmentation accuracy of the DCNN was similarly high between the internal (0.97±0.01) and external (0.96±0.03) test sets (p=0.28) and demonstrated robust segmentation performance on public testing (0.93±0.03). Agreement of liver volumetry was satisfactory in the internal (ICC, 0.99), external (ICC, 0.97), and publ...