Frontiers in Oncology, Год журнала: 2025, Номер 15
Опубликована: Апрель 16, 2025
Background/Aim This study develops a machine learning-based predictive model to evaluate the survival outcomes of elderly patients with unresectable hepatocellular carcinoma (HCC) undergoing radiotherapy. Methods The 2377 from SEER database were divided into training and internal validation cohorts. Additionally, 99 our hospital used for an external cohort. In cohort, 101 radiomics models developed, optimal model’s performance was subsequently evaluated in both Results StepCox + GBM demonstrated highest C-index 0.7 further using area under receiver operating characteristic (AUC-ROC) curves, AUC values ranging 0.736 0.783, indicating strong performance. Furthermore, calibration curve decision curves confirmed that had good Conclusions could help optimize use radiotherapy HCC patients, improving guiding personalized treatment strategies.
Язык: Английский