A novel nomogram combined with radiomics features and clinical features to predict pathological grading of bladder cancer

Background：Based on multi-parameter thin-slice enhanced CT texture features and related clinical indicators, a preoperative pathological grade prediction model of bladder urothelial carcinoma (GUC) was established.

Method：The CT images and clinical data of 372 patients with urothelial carcinoma in our hospital from January 2015 to October 2020 were collected. 372 patients were divided into two groups: HGUC(n=190) and LGUC(n=182). All patients were divided into 10 groups on average, of which 7 were used as training group (n=259) and the remaining 3 as verification group (n=113). Then, by using 3D-Slicer software from all enhanced CT images split out the interested area (ROI), we separately extracted 1223 texture feature from the tumor image data  based on first, second, high order and filtering characteristics. The correlation coefficient (ICC > 0.75) was used between classes and least absolute shrinkage selection operator (LASSO) regression feature selection. Secondly, the clinical predictive model was obtained by logistic regression analysis. Finally, the selected clinical key indicators and imaging features were mapped. In order to verify the predictive ability of the nomogram, conformance index (C-index), calibration curve, Receiver operator characteristic (ROC) curve and clinical decision curve (DCA) were used to test the nomogram.

Result：11 radiomics features were significantly correlated with the pathological grade of bladder cancer. After comparing the four models, logistic regression model was proved to be the best prediction ability (AUC=0.858). The results of multivariate analysis showed that age and proteinuria were independent risk factors. A comprehensive model for predicting the pathological grade of bladder cancer was constructed by combining clinical features with 11 imaging features. Compared with clinical feature model and imaging model, it was found that the predictive performance of imaging comprehensive model combined with clinical factors was the best (AUC=0.864).

Conclusion：The radiomics model has the best ability to predict high-or low-grade BCA (AUC=0.864).