Validating estimation of penile curvature
| Article DOI | https://doi.org/10.1016/j.jpurol.2023.09.008 |
| Objective | Assess the feasibility and validity of an AI-based model for the automatic measurement of severity of penile curvature |
| AI Approach | Segmentation |
| Data Source(s) | Seven 3D-printed penile models with various curvature angles; images taken of penile models with a smartphone |
| Model Input | 256 x 256 .jpeg images of penile models taken by a smartphone |
| Model Outcome | Performance of curvature estimation |
| Model Metrics | ICC 0.753; MAE 9.07 deg; Average Time: 22s |
| Model Usability | No available code or dataset. No publicly available tool. |
AI = Artificial intelligence, ICC = Intra-class correlation, AUC = Area under the receiver operator characteristic
AIPEDURO 
Leave a comment