Objective To explore the predictive value of radiomics nomogram based on multiparameter MRI for the status of human epidermal growth factor receptor 2(HER-2)in endometrial cancer(EC).Methods A total of 154 patients with pathologically proved EC were retrospectively selected and randomly divided into training set(108 cases)and validation set(46 cases)according to the ratio of 7∶3.Radiomics features were extracted from the preoperative multiparameter MRI images of the patients.The predictive performance of different machine learning algorithms was evaluated by receiver operating characteristic(ROC)curves,and the nomogram was constructed in combination with clinical risk factors.Results The degree of differentiation and depth of myometrial invasion were clinical risk factors for predicting HER-2 positivity in EC patients.The support vector machine(SVM)was selected as the best radiomics model.The nomogram showed the highest predictive performance in the training and validation sets,with area under the curve(AUC)of 0.926 and 0.897,respectively.Conclusion The radiomics nomogram based on multiparameter MRI can accurately predict the HER-2 status of EC patients before surgery and can be used to guide clinical treatment decisions.

Objective To explore the predictive value of radiomics nomogram based on multiparameter MRI for the status of human epidermal growth factor receptor 2(HER-2)in endometrial cancer(EC).Methods A total of 154 patients with pathologically proved EC were retrospectively selected and randomly divided into training set(108 cases)and validation set(46 cases)according to the ratio of 7∶3.Radiomics features were extracted from the preoperative multiparameter MRI images of the patients.The predictive performance of different machine learning algorithms was evaluated by receiver operating characteristic(ROC)curves,and the nomogram was constructed in combination with clinical risk factors.Results The degree of differentiation and depth of myometrial invasion were clinical risk factors for predicting HER-2 positivity in EC patients.The support vector machine(SVM)was selected as the best radiomics model.The nomogram showed the highest predictive performance in the training and validation sets,with area under the curve(AUC)of 0.926 and 0.897,respectively.Conclusion The radiomics nomogram based on multiparameter MRI can accurately predict the HER-2 status of EC patients before surgery and can be used to guide clinical treatment decisions.