Deep learning-assisted tumor radiomic dynamics on MRI predict pathological complete response in HCC undergoing immune-based therapy followed by hepatectomy

Key Points

• The study aims to evaluate how dynamic radiomic changes from MRI can predict pathological complete response (pCR) in patients with hepatocellular carcinoma (HCC) undergoing immune-based therapy.
• Analyzed pre- and post-treatment MRI data for uHCC patients.
• Utilized delta radiomics to assess changes in tumor characteristics.
• Combined delta radiomics with alpha-fetoprotein (AFP) response for enhanced prediction accuracy.
• Radiomic changes effectively forecasted pCR in patients after conversion therapy.
• Combining delta radiomics with AFP gave a significantly improved prediction performance.
• The approach offers a non-invasive way to assess treatment effectiveness.