Deep learning model-assisted detection of kidney stones on computed tomography
Problem
Detecting kidney stones of different sizes on unenhanced CT images across axial, coronal and other planes.
Why it matters
Kidney stones are common and accurate CT detection directly affects diagnosis and patient care.
What the AI did
A deep learning model evaluated 2,959 CT images from 455 patients (405 with stones) and its accuracy, sensitivity and specificity were compared against two radiologists' reviews.
Results
Training-group accuracy 97.3–99.1% across planes and stone-size groups; abstract truncated before full test-set metrics
Impact
High reported training accuracies suggest AI can aid stone detection, but test performance is not shown in the abstract.
Record
| Venue | International braz j urol |
|---|---|
| Published | 2022 |
| Work type | article |
| Authors | Alper Caglayan, Mustafa Ozan Horsanalı, Kenan Kocadurdu, Eren Ismailoglu, Serkan Güneyli |
| Citations | 94 (OpenAlex, at import time) |
| References | 30 |
| Model or method | deep learning model (specific architecture not stated in source) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1590/s1677-5538.ibju.2022.0132 |
| Open access | diamond |
| Keywords | deep learning, kidney stones, CT imaging, detection, radiology |
Importance rating
real but narrower advance
Kidney stone CT detection aids common diagnostic workflow; abstract claims reduced reporting time/cost and earlier diagnosis but gives no prevalence scale, and test-set accuracies were modest.