Tailored Prognostication with Deep Learning After HCC Resection

Using deep-learning models with CT imaging can significantly improve prognostication for patients after hepatocellular carcinoma surgery.

  • Postoperative recurrence rates within 2 and 5 years are 52.6% vs. 18.5% and 78.9% vs. 46.7% in high-risk vs. low-risk groups (p < 0.001).
  • 5-year mortality rates are 45.1% vs. 9.2% and 10-year rates are 87.1% vs. 43.2% (p < 0.001).

This tool helps tailor patient selection and optimize treatment plans, enhancing surgical outcomes.

  • Models achieved an AUC of 0.70 for 2 and 5-year recurrence and 0.80 for 5 and 10-year survival.

Journal Article by Shinkawa H, Ueda D (…) Ishizawa T et 14 al. in Hepatol Res

© 2025 Japan Society of Hepatology.

read the whole article in Hepatol Res

open it in PubMed