CT Radiomics–based Machine Learning to Identify Intratumoral Fibrosis and Underlying Angiogenesis in Hepatocellular Carcinoma

Key Points

• To investigate the effectiveness of Rad-ITF scores in classifying intratumoral fibrosis and its association with angiogenesis in hepatocellular carcinoma.
• Utilized CT radiomics and machine learning algorithms to analyze images.
• Correlated Rad-ITF scores with angiogenesis and prognosis in patients undergoing TACE.
• Rad-ITF scores effectively classified intratumoral fibrosis grades.
• Scores showed significant correlation with angiogenesis in both TCGA test and genomic test cohorts.