biostudies-literatureShu PChen Xiaoping Foundation for the development of science and technology of Hubei province217https://www.ebi.ac.uk/biostudies/studies/S-EPMC11954292biostudies-literatureUnknown30(1)<h4>Background</h4>Arteriovenous fistula stenosis is a common complication in hemodialysis patients, yet effective predictive tools are lacking. This study aims to develop an interpretable machine learning model for stenosis risk prediction.<h4>Methods</h4>Clinical data from 974 patients (55 features) undergoing arteriovenous fistula dialysis at The Central Hospital of Wuhan (2017-2024) were analyzed retrospectively. The dataset was split into training (70%) and test (30%) sets. Seven models-Random Forest, XGBoost, Support Vector Machine, Logistic Regression, K-Nearest Neighbors, Artificial Neural Network, and Decision Tree-were trained. Performance was evaluated using F1 score, accuracy, specificity, precision, recall, and AUC-ROC. SHAP values identified key predictors in the optimal model.<h4>Results</h4>XGBoost achieved the highest AUC (0.829, 95% CI 0.785-0.880). SHAP analysis highlighted seven critical predictors: number of surgeries, prothrombin time activity, lymphocyte count, fistula duration, triglycerides, vitamin B12, and C-reactive protein.<h4>Conclusion</h4>The XGBoost model effectively predicts arteriovenous fistula stenosis risk using clinical data. SHAP explanations enhance clinical interpretability, aiding personalized care strategies.European journal of medical researchMachine learning-based risk prediction model for arteriovenous fistula stenosis.PMC11954292CXPJJH124001-2404Qiu JShu PHuo SXu FHuang LWang XBai HfalseMachine learning-based risk prediction model for arteriovenous fistula stenosis.<h4>Background</h4>Arteriovenous fistula stenosis is a common complication in hemodialysis patients, yet effective predictive tools are lacking. This study aims to develop an interpretable machine learning model for stenosis risk prediction.<h4>Methods</h4>Clinical data from 974 patients (55 features) undergoing arteriovenous fistula dialysis at The Central Hospital of Wuhan (2017-2024) were analyzed retrospectively. The dataset was split into training (70%) and test (30%) sets. Seven models-Random Forest, XGBoost, Support Vector Machine, Logistic Regression, K-Nearest Neighbors, Artificial Neural Network, and Decision Tree-were trained. Performance was evaluated using F1 score, accuracy, specificity, precision, recall, and AUC-ROC. SHAP values identified key predictors in the optimal model.<h4>Results</h4>XGBoost achieved the highest AUC (0.829, 95% CI 0.785-0.880). SHAP analysis highlighted seven critical predictors: number of surgeries, prothrombin time activity, lymphocyte count, fistula duration, triglycerides, vitamin B12, and C-reactive protein.<h4>Conclusion</h4>The XGBoost model effectively predicts arteriovenous fistula stenosis risk using clinical data. SHAP explanations enhance clinical interpretability, aiding personalized care strategies.2025-01-01T00:00:00Z2025 Mar2025-07-13T03:04:06.816Z2025-07-13T03:04:06.816ZS-EPMC119542924015601610.1186/s40001-025-02490-x