Abstract: Objective: To explore the value of using radiomics based on contrast enhanced T1 weighted imaging(T1WI) in differentiating high-grade glioma recurrence from pseudoprogression. Methods: 104 patients with pathologically confirmed high-grade glioma were enrolled in this retrospective study, and they were divided into 71 cases in the recurrence group and 33 cases in the pseudoprogression group according to the secondary surgical pathology or response assessment in neuro-oncology (RANO). The cases were divided randomly into a training cohort and a validation cohort at a ratio of 7:3. Volume of interest (VOI) were delineated manually on contrast enhanced T1 weighted imaging. A total of 1648 radiomics features were extracted by FeAture Explorer software. Principal component analysis (PCA) and least absolute shrinkage and selection operator (LASSO) were used to process and select features. Support Vector Machine (SVM) was used to conduct the radiomics model. Area under the curve (AUC) and calibration curve were used to assess the efficacy of the model. Results: The radiomics model had an AUC of 0.929, accuracy of 0.889, sensitivity 72.7%, and specificity 100% in the training cohort to differentiate high-grade glioma recurrence from pseudoprogression. The validation cohort had an AUC of 0.853, accuracy of 0.813, sensitivity 90.9%, and specificity 71.4%. Calibration curve showed a good agreement between predicted and actual probabilities. Conclusion: The radiomics model based on contrast enhanced T1 weighted imaging showed great diagnostic performance and was promising in differentiating high-grade glioma recurrence from pseudoprogression.