AI-Driven Multimodal Approach: A Comprehensive Framework for Liver Tumor Diagnosis and Prognosis Using MRI

This review paper explores existing literature on Artificial Intelligence (AI) driven liver tumor diagnosis and prognosis using Magnetic Resonance Imaging (MRI) and proposes a comprehensive methodological framework. It examines segmentation, feature extraction, and predictive modeling techniques, integrating both convolutional and transformer based architectures. The study highlights challenges such as data scarcity and multimodal data fusion, offering a structured approach to improve classification accuracy for hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma (ICCA), and combined hepatocellular-cholangiocarcinoma (cHCC-CCA). By aiming to leverage pretrained models and hybrid learning strategies on our available data, this framework aims to enhance diagnostic reliability and inform future advancements in AI-assisted liver cancer diagnosis and prognosis.