Development of machine learning models to predict the risk of fungal infection following flexible ureteroscopy lithotripsy DOI Creative Commons

Haofang Zhang,

Changbao Xu,

Chenge Hu

et al.

BMC Medical Informatics and Decision Making, Journal Year: 2025, Volume and Issue: 25(1)

Published: April 10, 2025

The flexible ureteroscopy lithotripsy (F-URL) is an important treatment for upper urinary tract stones. However, urolithiasis, surgical procedures, and catheter placement are risk factors fungal infections. Our study aimed to construct a machine learning algorithm predictive model predict the of infection following F-URL. This retrospectively collected clinical data patients who underwent F-URL at Second Affiliated Hospital Zhengzhou University from January 2016 March 2024. were divided into non-fungal group based on whether occurred within three months post-surgery. patient December 2023 used as training data, 2024 testing set. was randomly set validation ratio 90:10. Use LASSO regression screen features Nine algorithms, Logistic Regression (LR), k-Nearest Neighbours (KNN), Support Vector Machines (SVM), Random Forest (RF), Categorical Boosting (CatBoost), eXtreme Gradient (XGBoost), Adaptive (AdaBoost), (GBM), Neural Network (NNet), models. performance these nine models evaluated best selected set, evaluate model's generalization ability using Visualize constructed optimal SHapley additive interpretation (SHAP) value method. SHAP force plots established show application prediction individual level. A total 13 models: age, diabetes mellitus (DM), history malignancy, being bedridden, admission white blood cells (WBC), preoperative ureteral stenting, operation time, postoperative fever, Neu, carbapenem antibiotics use, duration antibiotic therapy, length hospital stay (LOS), stent duration. Comparing 9 models, we found that XGBoost had performance. shows good discrimination, applicability in developed this has evaluating

Language: Английский

Development of machine learning models to predict the risk of fungal infection following flexible ureteroscopy lithotripsy DOI Creative Commons

Haofang Zhang,

Changbao Xu,

Chenge Hu

et al.

BMC Medical Informatics and Decision Making, Journal Year: 2025, Volume and Issue: 25(1)

Published: April 10, 2025

The flexible ureteroscopy lithotripsy (F-URL) is an important treatment for upper urinary tract stones. However, urolithiasis, surgical procedures, and catheter placement are risk factors fungal infections. Our study aimed to construct a machine learning algorithm predictive model predict the of infection following F-URL. This retrospectively collected clinical data patients who underwent F-URL at Second Affiliated Hospital Zhengzhou University from January 2016 March 2024. were divided into non-fungal group based on whether occurred within three months post-surgery. patient December 2023 used as training data, 2024 testing set. was randomly set validation ratio 90:10. Use LASSO regression screen features Nine algorithms, Logistic Regression (LR), k-Nearest Neighbours (KNN), Support Vector Machines (SVM), Random Forest (RF), Categorical Boosting (CatBoost), eXtreme Gradient (XGBoost), Adaptive (AdaBoost), (GBM), Neural Network (NNet), models. performance these nine models evaluated best selected set, evaluate model's generalization ability using Visualize constructed optimal SHapley additive interpretation (SHAP) value method. SHAP force plots established show application prediction individual level. A total 13 models: age, diabetes mellitus (DM), history malignancy, being bedridden, admission white blood cells (WBC), preoperative ureteral stenting, operation time, postoperative fever, Neu, carbapenem antibiotics use, duration antibiotic therapy, length hospital stay (LOS), stent duration. Comparing 9 models, we found that XGBoost had performance. shows good discrimination, applicability in developed this has evaluating

Language: Английский

Citations

0