Objective: To investigate the survival prediction probability of the nomogram from retro-EMBRACE and the T-score in patients with locally advanced cervical cancer (LACC). Materials and Methods: A total of 204 patients with LACC who underwent curative chemoradiotherapy and brachytherapy (BT) between 2010 and 2021 were included in our single-center retrospective study. Clinical records, examinations, and magnetic resonance images (MRI) before and after external beam radiotherapy (EBRT) were retrospectively reviewed to obtain information on age, tumor size, parametrial involvement, ureteral involvement, bladder invasion, uterine involvement, high-risk clinical target volume at the first brachytherapy application, lymph node involvement, vaginal involvement, recurrence, metastasis, and last follow-up. The 5-year overall survival probabilities of the patients were determined by nomogram. T-score was calculated separately at diagnosis (TS_D) and brachytherapy (TS_BT), and their effects on local recurrence-free survival, disease-free survival, and overall survival were analyzed. Results: The median age was 52 (29–89). The 5-year survival rate of the patients was calculated to be 90.18%. The median nomogram’s survival estimate for 60 months was 70.35% (20.9–87.1). The median TS_D and TS_BT were 5.5 (1–16) and 1 (0–6), respectively. According to the multivariate Cox regression models, TS_D (HR = 1.203, 95% CI 1.021–1.417, p = 0.024) was significantly associated with local recurrence-free survival. Conclusions: This study demonstrated that the nomogram’s predictions for 60-month overall survival are underestimates. Prognosis can be estimated using the TS_D, which can be easily obtained with a clinical examination and detailed MRI examination.

Background: This study aims to evaluate the contribution of clinical and radiomic features to machine learning-based models for survival prediction in patients with locally advanced cervical cancer. Methods: Clinical and radiomic data from 161 patients were retrospectively collected from a single center. Radiomic features were obtained from contrast-enhanced magnetic resonance imaging (MRI) T1-weighted (T1W), T2-weighted (T2W), and diffusion-weighted (DWI) sequences. After data cleaning, feature engineering, and scaling, survival prediction models were created using the CatBoost algorithm with different data combinations (clinical, clinical + T1W, clinical + T2W, clinical + DWI). The performance of the models was evaluated using test accuracy, precision, recall, F1-score, ROC curve, and Bland–Altman analysis. Results: Models using both clinical and radiomic features showed significant improvements in accuracy and F1-score compared to models based solely on clinical data. In particular, the CatBoost_CLI + T2W_DMFS model achieved the best performance, with a test accuracy of 92.31% and an F1-score of 88.62 for distant metastasis-free survival prediction. ROC and Bland–Altman analyses further demonstrated that this model has high discriminative power and prediction consistency. Conclusions: The CatBoost algorithm shows high accuracy and reliability for survival prediction in locally advanced cervical cancer when clinical and radiomic features are combined. The addition of radiomics data significantly improves model performance.