Deep learning model for assessing survival benefits in hepatocellular carcinoma patients undergoing intra-arterial therapies based on proliferative subtype - Takeaways - MDSpire

Deep learning model for assessing survival benefits in hepatocellular carcinoma patients undergoing intra-arterial therapies based on proliferative subtype

  • By

  • Fei Cao

  • Chao An

  • Shaolong Li

  • Xiaochun Hu

  • Da Li

  • Jiayu Pan

  • Zhijun Geng

  • Fei Gao

  • Mengxuan Zuo

  • November 19, 2025

  • 0 min

Share

  • 1

    A multitask deep learning system was developed to detect proliferative hepatocellular carcinoma (HCC) and predict survival after intra-arterial therapy.

  • 2

    The study analyzed contrast-enhanced CT scans from 2147 patients, achieving AUCs of 0.825 and 0.792 for detecting proliferative HCC.

  • 3

    Prognostic nomograms combining radiomic and clinical variables outperformed traditional staging systems in predicting survival outcomes.

  • 4

    High-risk patients showed significant survival benefits from hepatic arterial infusion chemotherapy compared to transarterial chemoembolization.

  • 5

    This non-invasive deep learning approach supports personalized treatment choices for unresectable HCC, potentially improving patient outcomes.

Original Source(s)

Deep learning model for assessing survival benefits in hepatocellular carcinoma patients undergoing intra-arterial therapies based on proliferative subtype

  • by Fei Cao, Chao An, Shaolong Li, Xiaochun Hu, Da Li, Jiayu Pan, Zhijun Geng, Fei Gao, Mengxuan Zuo

  • November 19, 2025

Related Content

HemaContour: explicit parametric contour learning for robust ICH segmentation on non-contrast CT

Deep multimodal state-space fusion of endoscopic-radiomic and clinical data for survival prediction in colorectal cancer

STD-Net: a spatio-temporal decoupling network for multiphasic liver lesion segmentation and characterization