Dilating VUR from ultrasound, VUR + UTI.
| Article DOI | https://doi.org/10.1016/j.jpurol.2023.11.003 |
| Objective | To reliably predict dilating VUR from early postnatal US in patients with hydronephrosis |
| AI Approach | Optimal Classification Tree |
| Data Source(s) | Single institutional series (280 patients, 530 renal units) |
| Model Input | Patient demographics (age at US, sex, kidney laterality) and UTD features (primary antero-posterior diameter, central calyceal dilation, peripheral calyceal dilation, parenchymal thickness, parenchymal appearance, distal ureteral dilation) |
| Model Outcome | Grade 4/5 VUR |
| Model Metrics | AUC 0.81 |
| Model Usability | Final prediction model uses a tree structure that is available for clinical use. |
AI = Artificial intelligence, VUR = Voiding cystourethrogram, US = Ultrasound, AUC = Area under the receiver operator characteristic, UTD = Upper tract dilation
AIPEDURO 
Leave a comment