Abnormal versus normal kidney, Hydronephrosis.
| Article DOI | https://doi.org/10.2196/40878 |
| Objective | To evaluate the diagnostic performance of deep learning techniques in classifying kidney images as normal or abnormal |
| AI Approach | CNN with transfer learning |
| Data Source(s) | Single institutional series (1599 images) |
| Model Input | Renal US images (224×224 pixels) |
| Model Outcome | Abnormal vs. normal |
| Model Metrics | Accuracy: 93% AUC: 0.96 |
| Model Usability | NA |
AI = Artificial intelligence, UPJO = Ureteropelvic junction obstruction, CNN = Convolutional Neural Network, AUC = Area-under-the-receiver-operator-characteristic
AIPEDURO 
Leave a comment