A combined model based on CT radiomics and clinical variables to predict uric acid calculi

Abstract Introduction: The aim of this study is to develop and validate a CT-based radiomics and clinical variables diagnostic model for preoperative prediction of uric acid in urinary stones. Methods: In this retrospective study, 370 patients with urolithiasis who underwent preoperative urinary CT scan were enrolled. The CT images of each patient were manually segmented and radiomics features were extracted. Sixteen radiomics features were selected. Logistic regression (LR), random forest (RF) and support vector machine (SVM) were used to model the selected features, and the model with the best performance was selected. Multivariate logistic regression was used to screen out significant clinical variables, the radiomics features and clinical variables were combined to construct a nomogram model. The area under the receiver operating characteristic (ROC) curve (AUC), etc. were used to evaluate the diagnostic performance of the model. Result: Among the three machine learning models, the LR model had the best result on the training set (AUC=0.855), and the performance in the training set and test set was similar (AUC=0.837). Therefore, the LR model was finally used to construct the nomogram. The AUC of the nomogram model in the training set and validation set were 0.870 and 0.837, respectively, which were significantly higher than those of the radiomics model and the clinical feature model. Conclusion: The radiomics model based on CT has a good performance in distinguishing uric acid stones from non-uric acid stones, and the nomogram model has the best diagnostic performance among the three models, which can provide effective reference for clinical decision-making.