Neuro / Head & NeckAI / InformaticsResearch
MRI and Radiotherapy Dose AI Personalizes Follow-Up After De-Intensified Nasopharyngeal Carcinoma Treatment
International journal of radiation oncology, biology, physicsyesterday
AI-guided risk-adapted follow-up after de-intensified treatment for stage II nasopharyngeal carcinoma reduced follow-up visits for >90% of failure-free patients (no more than six visits over five years), with external AUC 0.986 and high sensitivity.
- In a target trial emulation, omitting concurrent chemotherapy yielded comparable survival, establishing a safely de-intensified baseline.
- The multimodal AI (MRI + radiotherapy dose maps) predicted treatment failure timing with AUC 0.991 internally and 0.986 in external validation across 2,148 patients from five centers.
- The AI strategy recommended only up to six visits in five years for high-risk patients, maintaining high sensitivity for detecting failures.
Automated summary
RadPigeon summaries are original and for information only. They are not clinical advice.