1. Deep learning-based precise prediction and early detection of radiation-induced temporal lobe injury for nasopharyngeal carcinomaResearch in context
- Author
-
Pu-Yun OuYang, Bao-Yu Zhang, Jian-Gui Guo, Jia-Ni Liu, Jiajian Li, Qing-He Peng, Shan-Shan Yang, Yun He, Zhi-Qiao Liu, Ya-Nan Zhao, Anwei Li, Yi-Shan Wu, Xue-Feng Hu, Chen Chen, Fei Han, Kai-Yun You, and Fang-Yun Xie
- Subjects
Deep learning ,Early detection ,Nasopharyngeal carcinoma ,Planning evaluation ,Radiation-induced temporal lobe injury ,Medicine (General) ,R5-920 - Abstract
Summary: Background: Radiotherapy is the mainstay of treatment for nasopharyngeal carcinoma. Radiation-induced temporal lobe injury (TLI) can regress or resolve in the early phase, but it is irreversible at a later stage. However, no study has proposed a risk-based follow-up schedule for its early detection. Planning evaluation is difficult when dose-volume histogram (DVH) parameters are similar and optimization is terminated. Methods: This multicenter retrospective study included 6065 patients between 2014 and 2018. A 3D ResNet-based deep learning model was developed in training and validation cohorts and independently tested using concordance index in internal and external test cohorts. Accordingly, the patients were stratified into risk groups, and the model-predicted risks were used to develop risk-based follow-up schedules. The schedule was compared with the Radiation Therapy Oncology Group (RTOG) recommendation (every 3 months during the first 2 years and every 6 months in 3–5 years). Additionally, the model was used to evaluate plans with similar DVH parameters. Findings: Our model achieved concordance indexes of 0.831, 0.818, and 0.804, respectively, which outperformed conventional prediction models (all P
- Published
- 2023
- Full Text
- View/download PDF