Abstract #2336

AI Methods for Predicting Sensitivity of Total Neoadjuvant Treatment (TNT) in Rectal Cancer Based on Multiparameter MRI and Clinical Data

Ganlu Ouyang¹, Zhebin Chen^2,3, Jitao Zhou¹, Meng Dou^2,3, Xu Luo^2,3, Han Wen^2,3, Yu Yao^2,3, and Xin Wang¹

¹Department of Radiation Oncology, Cancer Center, West China Hospital, Sichuan University, Chengdu, China, ²Chengdu Institute of Computer Application, Chinese Academy of Sciences, Chengdu, China, ³University of Chinese Academy of Sciences, Beijing, China

This work aimed for developing a model for predicting sensitivity and response of total neoadjuvant treatment (TNT) for locally advanced rectal cancer (LARC) based on baseline magnetic resonance imaging (MRI) and clinical data by artificial intelligence method. The results showed that the models for predicting high sensitivity and pCR built with radiomics features achieved the mean area under the ROC curve (AUC) of 0.85 respectively, while the other built with deep-learning (DL) method yielded the mean AUC of 0.82 and 0.84 respectively. The models of two methods for predicting high sensitivity and pCR may be valuable in clinical practice.

This abstract and the presentation materials are available to members only; a login is required.

Join Here