A Machine Learning Model for Predicting Prognosis in HCC Patients With Diabetes After TACE

• DOI: 10.2147/jhc.s496481
• Publication Date: 2025-01-20T16:15:12Z
• Authors: Linxia Wu; Lei Chen; Lijie Zhang; Yiming Liu; Die Ouyang; Wenlong Wu; Yu Lei; Ping Han; Huangxuan Zhao; Chuansheng Zheng
• Keywords: shapley additive explanations.; type ii diabetes mellitus; overall survival; Neoplasms. Tumors. Oncology. Including cancer and carcinogens; hepatocellular carcinoma; transarterial chemoembolization; RC254-282; Original Research

Purpose: Type II diabetes mellitus (T2DM) has been found to increase the mortality of patients with hepatocellular carcinoma (HCC).Therefore, this study aimed establish and validate a machine learning-based explainable prediction model prognosis in HCC T2DM undergoing transarterial chemoembolization (TACE).Patients Methods: The was developed using data from derivation cohort comprising three medical centers, followed by external validation utilizing patient extracted another center.Further, five predictive models were employed for 1-, 2-, 3-year survival, respectively.Prediction performance assessed receiver operating characteristic (ROC), calibration, decision curve analysis curves.Lastly, SHapley Additive exPlanations (SHAP) method used interpret final ML model.Results: A total 636 included.Thirteen variables selected development.The random survival forest (RSF) exhibited excellent internal cohort, areas under ROC (AUCs) 0.824, 0.853, 0.810 groups, respectively.This also demonstrated remarkable discrimination achieving AUCs 0.862, 0.815, 0.798 respectively.SHAP summary plots created RSF model.Conclusion: An good incorporating simple parameters.This prognostic may assist physicians early clinical intervention improve outcomes comorbid after TACE.