Key points are not available for this paper at this time.
Cervical cancer (CC) continues to impose a substantial global health burden and remains one of the most prevalent malignancies among women worldwide. Radiotherapy is a cornerstone treatment for locally advanced disease, and its precision critically impacts tumor control and treatment-related toxicity. Within the evolving paradigm of precision oncology, radiomics and artificial intelligence (AI) have emerged as promising tools to personalize radiotherapy by improving target delineation, predicting treatment response, refining prognostic stratification, and facilitating individualized toxicity risk assessment. This narrative review synthesizes and critically appraises the current evidence on the application of radiomics and AI in CC radiotherapy, focusing on three principal domains: automated target volume delineation, prediction of prognosis and treatment response, and forecasting of radiotherapy-induced toxicities. We further evaluate the methodological rigor and translational readiness of existing studies. Despite encouraging technical performance, most available evidence remains retrospective, with limited prospective validation and uncertain impact on clinical decision-making. Clinical implementation is further challenged by imaging heterogeneity, insufficient standardization, and limited model interpretability. Future research should prioritize large-scale multicenter validation, methodological standardization, and prospective evaluation to determine whether radiomics-guided strategies can meaningfully improve patient outcomes and support integration into routine clinical practice.
He et al. (Tue,) studied this question.