Objective To establish the predictive model for the efficacy of neoadjuvant chemoradiothera-py(nCRT)based on the multi-sequence magnetic resonance imaging(MRI)images before and after nCRT.Methods The clinical and pathological data of patients with locally advanced rectal cancer(T3-4N+M0)who underwent nCRT at Gansu Provincial People's Hospital from January 2020 to May 2023 were retrospectively analyzed.According to the efficacy of nCRT,the patients were divided into remission group and non-remission group,and the multi-sequence imaging features of MRI were analyzed.At the same time,the single-sequence imaging model and multi-sequence combined imaging model were construc-ted,and the predictive effect of the single-sequence imaging model and multi-sequence combined imaging model on the efficacy of nCRT was evaluated by using the receiver operating characteristic(ROC)curve.Decision curve analysis(DCA)was used to evaluate the clinical application value of each model.Results The area under the curve(AUC)of the T2WI+DWI multi-sequence combined imaging model was 0.919,which was significantly higher than that of the T2WI+DWI single-sequence imaging model.The calibration ability and discriminatory ability of the multi-sequence combined imaging predictive model were better than those of the single-sequence imaging model,with a sensitivity of 0.958 and a specificity of 0.800,and a 95％confidence interval(0.79,1.00).Conclusion The predictive model based on multi-se-quence combined imaging is superior to the single-sequence imaging model in predicting the efficacy of nCRT.
Magnetic resonance imagingRectal cancerNeoadjuvant chemoradiotherapyTumor responsePredictive model
NETL
NSTL
万方数据
