Deep learning LI-RADS grading system based on contrast enhanced multiphase MRI for differentiation between LR-3 and LR-4/LR-5 liver tumors

Abstract

Background: To develop a deep learning (DL) method based on multiphase, contrast-enhanced (CE) magnetic resonance imaging (MRI) to distinguish Liver Imaging Reporting and Data System (LI-RADS) grade 3 (LR-3) liver tumors from combined higher-grades 4 and 5 (LR-4/LR-5) tumors for hepatocellular carcinoma (HCC) diagnosis.Methods: A total of 89 untreated LI-RADS-graded liver tumors (35 LR-3, 14 LR-4, and 40 LR-5) were identified based on the radiology MRI interpretation reports. Multiphase 3D T1-weighted gradient echo imaging was acquired at six time points: pre-contrast, four phases immediately post-contrast, and one hepatobiliary phase after intravenous injection of gadoxetate disodium. Image co-registration was performed across all phases on the center tumor slice to correct motion. A rectangular tumor box centered on the tumor area was drawn to extract subset tumor images for each imaging phase, which were used as the inputs to a convolutional neural network (CNN). The pre-trained AlexNet CNN model underwent transfer learning using liver MRI data for LI-RADS tumor grade classification. The output probability number closer to 1 or 0 indicated a higher possibility of being combined LR-4/LR-5 tumor or LR-3 tumor, respectively. Five-fold cross validation was used for training (60% dataset), validation (20%) and testing processes (20%).Results: The DL CNN model for LI-RADS grading using inputs of multiphase liver MRI data acquired at three time points (pre-contrast, arterial, and washout phase) achieved a high accuracy of 0.90, sensitivity of 1.0, precision of 0.835, and AUC of 0.95 with reference to the expert human radiologist report. The CNN output of probability provided radiologists a confidence level of the model's grading for each liver lesion.Conclusions: An AlexNet CNN model for LI-RADS grading of liver lesions provided diagnostic performance comparable to radiologists and offered valuable clinical guidance for differentiating intermediate LR-3 liver lesions from more-likely malignant LR-4/LR-5 lesions in HCC diagnosis.

Progress of MRI Radiomics in Hepatocellular Carcinoma

“…In recent years, DL technology has been developed and has achieved excellent performance in the classification of hepatic lesions ( 65 – 71 ). Hamm CA et al ( 65 ) developed a proof-of-concept convolutional neural network (CNN)-based DL system and classified 494 hepatic lesions from six categories on MRI.…”

State of the Art in Artificial Intelligence and Radiomics in Hepatocellular Carcinoma

“…Hamm et al [ 43 ] employed a CNN to classify liver lesions on MRI, with an accuracy of 92%, a sensitivity of 92% and a specificity of 98% compared to a sensitivity and specificity of 82.5% and 96.5% obtained on these same cases by radiologists. Wu et al [ 44 ] also investigated the usefulness of a CNN model for LI-RADS grading using a multiphase liver MRI. Specifically, the outcome of interest was the differentiation between LR-3 and LR-4/LR-5 tumors, achieving an accuracy of 90%, sensitivity of 100% and AUC of 0.95.…”

An Overview of Artificial Intelligence Applications in Liver and Pancreatic Imaging

“…Referring to HCC in a clinical setting, another paper evaluated the performance of DL applied to the Liver Imaging Reporting and Data System (LI-RADS) [ 42 ]. This model, based on MR images, reaches a correct classification rate of 90% and an AUC of 0.95 compared to radiologists in the differentiation between LI-RADS 3 and LI-RADS 4–5 lesions [ 43 ].…”