Infections after pyeloplasty
| Article DOI | https://dx.doi.org/10.1038/s41598-024-82282-1 |
| Objective | To enhance predictive accuracy for urinary tract infections post-pediatric pyeloplasty |
| AI Approach | Logistic Regression, SVM, Random Forest, XGBoost, LightGBM and deep learning TabNet model |
| Data Source(s) | Single institutional series (764 patients) |
| Model Input | 25 clinical features including patient demographics, medical history, surgical details, and various postoperative indicators |
| Model Outcome | UTI after pyeloplasty |
| Model Metrics | LightGBM: AUC 0.78, accuracy 71%, sensitivity 0.71 Ensemble approach (LightGBM and TabNet): AUC 0.80, accuracy 80%, sensitivity 0.65 |
| Model Usability | De-identified datasets provided as a supplementary document. No usable code or application. |
AI = Artificial intelligence, SVM = Support vector machines, AUC = Area under the receiver operator characteristic
AIPEDURO 
Leave a comment