Predicting Microvascular Invasion and Recurrence-Free Survival in Hepatocellular Carcinoma Using DCE-MRI Habitat Analysis and Deep Learning

Motivation: Hepatocellular carcinoma (HCC) has high incidence and mortality rates, with microvascular invasion (MVI) being a key factor in recurrence and metastasis. Goal(s): To develop a model combining habitat analysis and deep learning (DL) features from DCE-MRI to predict MVI and recurrence-free survival (RFS) in HCC patients. Approach: The study involved retrospective analysis of DCE-MRI data from HCC patients, applying clustering methods and extracting radiomic and DL features to build predictive models. Results: The fusion model combining habitat risk scores (HRS) and DL features achieved the highest accuracy in predicting MVI and demonstrated robust performance in RFS risk stratification. Impact: The fusion model's high accuracy in predicting MVI and RFS can significantly enhance preoperative planning for HCC patients, guiding personalized treatment strategies and improving prognosis. This approach opens new avenues for non-invasive cancer diagnostics and risk stratification.