Deep learning model predicts response to TACE/TKI in HBV-related HCC with AUC 0.85

A deep learning model (ResNet-50) predicted objective response to TACE+TKI therapy in HBV-related unresectable HCC with AUC 0.85 (95% CI 0.75-0.95), outperforming clinical (0.59) and radiomics (0.71) models, and significantly stratified progression-free survival (P=0.011). Retro…

• Retrospective study of 243 patients with HBV-related uHCC, partitioned into training (clinical n=168; radiomics n=106) and test set (n=75).
• DL model was the only framework to significantly stratify PFS (P=0.011), and Grad-CAM revealed focal tumor activation in responders versus multifocal peripheral activation in non-responders.
• Single-center retrospective design without external validation; limited generalizability.

RadPigeon summaries are original and for information only. They are not clinical advice.