Survival prediction in colorectal cancer liver metastases using machine learning with SHAP-based interpretation - Takeaways - MDSpire

Survival prediction in colorectal cancer liver metastases using machine learning with SHAP-based interpretation

  • By

  • Nan Li

  • Baoxin Dong

  • Yu Liang

  • Likun Liu

  • Xixing Wang

  • Ce Zhang

  • Shulan Hao

  • June 10, 2026

  • 0 min

Share

  • 1

    The study developed an interpretable machine learning model to predict survival outcomes in colorectal cancer liver metastasis patients.

  • 2

    The optimized XGBoost algorithm achieved an AUC of 0.891 for 36-month survival prediction in the training cohort.

  • 3

    SHAP analysis identified TNM stage, liver metastasis burden, and TCM intervention intensity as key prognostic factors.

  • 4

    Traditional Chinese Medicine showed a protective association with survival probability in a dose-dependent manner.

  • 5

    The model was successfully translated into a web-based tool for personalized prognostic assessment and treatment planning.

Original Source(s)

Survival prediction in colorectal cancer liver metastases using machine learning with SHAP-based interpretation

  • by Nan Li, Baoxin Dong, Yu Liang, Likun Liu, Xixing Wang, Ce Zhang, Shulan Hao

  • June 10, 2026

Related Content

Research on gastrointestinal polyp detection method based on improved YOLOv7

Synergistic Immunomodulation: Integrating Physical Ablation with Immune Checkpoint Inhibitors for Metastatic Colorectal Carcinoma

Effectiveness of nurse-led dietary modification recommendations in the management of defecation dysfunction among patients with rectal cancer after sphincter-saving surgery: a scoping review