Development And Validation Of A CT-based Radiomics Nomogram For Preoperative Prediction Of Tumor Histologic Grade In Gastric Adenocarcinoma

Objectives: To develop and validate a radiomics nomogram to preoperatively predict the tumor histologic grade in gastric adenocarcinoma(GA).Methods: This retrospective study enrolled 592 patients with clinicopathologically confirmed gastric adenocarcinoma(low-grade: n = 154;high-grade: n = 438)from January 2008 to March 2018 who were divided into training(n = 450)and validation(n= 142)sets according to the time of computed tomography(CT)examination.Radiomic features were extracted from the portal venous–phase abdominal CT images.The MannWhitney U test and the least absolute shrinkage and selection operator(LASSO)regression model were used for feature selection and for data dimension reduction.Radiomics signature was developed based on the selected features.The radiomics signature and independent clinicopathologic risk factors were analyzed by multivariable logistic regression model,incorporated and finally presented as a radiomics nomogram.The performance of the nomogram was assessed concerning its calibration and discrimination.Results: A radiomics signature containing 12 selected features was associated with the histologic grade of GA with statistical significance(P < 0.05 for both training and validation sets).A nomogram including the radiomics signature and tumor location as predictors was developed.The model showed satisfied discrimination(AUC in the training set,0.752 [95% CI: 0.701 to 0.803];AUC in the validation set,0.793 [95% CI:0.711 to 0.874])and good calibration.Conclusion: This study developed a radiomics nomogram that incorporates tumor location and radiomics signatures,which could be useful in facilitating the preoperative individualized prediction of the histologic grade of GA.