AUTHOR=Qiu Junliang , Yan Minbo , Wang Haojie , Liu Zicheng , Wang Guojie , Wu Xianbo , Gao Qindong , Hu Hongji , Chen Junyong , Dai Yingbo 

TITLE=Identifying ureteral stent encrustation using machine learning based on CT radiomics features: a bicentric study

JOURNAL=Frontiers in Medicine

VOLUME=10

YEAR=2023

URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2023.1202486

DOI=10.3389/fmed.2023.1202486

ISSN=2296-858X

ABSTRACT=<sec id="sec1"><title>Obstructive</title><p>To develop and validate radiomics and machine learning models for identifying encrusted stents and compare their recognition performance with multiple metrics.</p></sec><sec id="sec2"><title>Methods</title><p>A total of 354 patients with ureteral stent placement were enrolled from two medical institutions and divided into the training cohort (<italic>n</italic> = 189), internal validation cohort (<italic>n</italic> = 81) and external validation cohort (<italic>n</italic> = 84). Based on features selected by Wilcoxon test, Spearman Correlation Analysis and least absolute shrinkage and selection operator (LASSO) regression algorithm, six machine learning models based on radiomics features were established with six classifiers (LR, DT, SVM, RF, XGBoost, KNN). After comparison with those models, the most robust model was selected. Considering its feature importance as radscore, the combined model and a nomogram were constructed by incorporating indwelling time. Accuracy, sensitivity, specificity, area under the curve (AUC), decision curve analysis (DCA) and calibration curve were used to evaluate the recognition performance of models.</p></sec><sec id="sec3"><title>Results</title><p>1,409 radiomics features were extracted from 641 volumes of interest (VOIs) and 20 significant radiomics features were selected. Considering the superior performance (AUC 0.810, 95%CI, 0.722–0.888) in the external validation cohort, feature importance of XGBoost was used as a radscore, constructing a combined model and a nomogram with indwelling time. The accuracy, sensitivity, specificity and AUC of the combined model were 98, 100, 97.3% and 0.999 for the training cohort, 83.3, 80, 84.5% and 0.867 for the internal cohort and 78.2, 76.3, 78.8% and 0.820 for the external cohort, respectively. DCA indicates the favorable clinical utility of models.</p></sec><sec id="sec4"><title>Conclusion</title><p>Machine learning model based on radiomics features enables to identify ureteral stent encrustation with high accuracy.</p></sec>