Decoding heterogeneity in solitary HCC: OATP-dependent HBP-MRI radiomics predicts microvascular invasion & prognosis.

Motivation: HCC is deadly, and MVI impacts prognosis. Goal(s): This study integrates multiple machine learning methods to predict MVI in HCC, revealing an ECM-OATP1B3 correlation and enhancing the biological interpretability of the radiomics model for prognostic assessment. Approach: Machine learning methods to identify radiomic features and VOI regions. RNA sequencing explored biological differences between two groups. The model's efficacy was validated through immunohistochemistry and staining. Results: The model identified optimal methods and margin regions, with AUC values of 0.80, 0.76, and 0.74. It stratified patients into two groups with significant survival differences. RNA sequencing revealed ECM-related genes and pathways. Immunohistochemistry validated the model's reliability. Impact: This study integrates multiple machine learning methods to predict MVI in HCC, revealing an ECM-OATP1B3 correlation and enhancing the biological interpretability of the radiomics model for prognostic assessment.