Determination of urinary stone composition using computed tomography using artificial intelligence.
- 1Department of Urology, School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile.
- 2Department of Urology, Hospital Sotero del Río, Santiago, Chile.
- 3School of Medicine, Pontificia Universidad Católica de Chile, Santiago, Chile.
Abstract
Introduction
Urinary stone composition analysis is essential for guiding treatment and preventive strategies, but conventional methods such as Fourier-transform infrared spectroscopy (FTIR) are costly and limited in availability. Artificial intelligence (AI), particularly convolutional neural networks (CNNs), offers potential as a non-invasive alternative using computed tomography (CT) imaging. The aim of the study was to evaluate the performance of a CNN model in predicting urinary stone composition from non-contrast CT (NCCT) images.Material and methods
A retrospective analysis was conducted using an institutional database of patients with urinary stones surgically removed between 2020 and 2024. Stone composition was confirmed via FTIR. A total of 1,487 NCCT scan images were used for CNN development and internal validation, with 105 scan images reserved for validation. Transfer learning with VGG16 architecture was applied, incorporating data augmentation and fine-tuning strategies. Performance metrics included accuracy, area under the ROC curve (AUC), and Cohen's κ coefficient.Results
The dataset comprised calcium oxalate monohydrate (74.9%), calcium oxalate dihydrate (17.6%), uric acid (2.7%), struvite (2.3%), and other compositions (2.5%). The CNN achieved an internal accuracy of 89.8% (AUC 0.92). In independent temporal validation the cohort accuracy was 85.4%, with an AUC of 0.81 and a Cohen's κ of 0.864. Misclassifications occurred mainly between calcium oxalate subtypes due to overlapping radiological features.Conclusions
CNN-based analysis of NCCT images provides accurate, non-invasive prediction of urinary stone composition, with performance comparable to conventional methods. This approach may improve preoperative planning and broaden diagnostic access, particularly in settings where FTIR is unavailable. Prospective multicenter studies are warranted to validate clinical implementation.
Related
- The effect of artificial intelligence-driven, computed tomography-based assessment of body composition parameters on 28-day mortality of patients with severe fever with thrombocytopenia syndrome: a retrospective cohort study.
- The integration of artificial intelligence into computed tomography: from segmentation to the standardisation of clinical decisions
- Association of sarcopenia with treatment tolerance in older adults with acute myeloid leukemia using artificial intelligence computed tomography–segmentation and a geriatric assessment
- Accuracy of artificial intelligence-based detection of dental crowding and spacing on cone-beam computed tomography.
- Artificial Intelligence-Derived Computed Tomography Phenotyping for Cardiovascular Risk Stratification in Transcatheter Mitral Valve Replacement.
BibTeX
@article{gallegos2026determination,
title = {{Determination of urinary stone composition using computed tomography using artificial intelligence.}},
author = {Hector Gallegos and Alejandro Rojas and Enzo Castiglioni and Carolina Verdugo and Benjamin Barraza and Benjamin Vargas and Stefano Brusoni and Gastón Astroza},
journal = {Central European journal of urology},
year = {2026},
month = jun,
doi = {10.5173/ceju.2025.0186},
url = {https://europepmc.org/article/MED/42904101},
note = {CC BY-NC-SA},
}
RIS
TY - JOUR
TI - Determination of urinary stone composition using computed tomography using artificial intelligence.
AU - Hector Gallegos
AU - Alejandro Rojas
AU - Enzo Castiglioni
AU - Carolina Verdugo
AU - Benjamin Barraza
AU - Benjamin Vargas
AU - Stefano Brusoni
AU - Gastón Astroza
PY - 2026
DA - 2026/06/18
JO - Central European journal of urology
DO - 10.5173/ceju.2025.0186
UR - https://europepmc.org/article/MED/42904101
AB - Introduction Urinary stone composition analysis is essential for guiding treatment and preventive strategies, but conventional methods such as Fourier-transform infrared spectroscopy (FTIR) are costly and limited in availability. Artificial intelligence (AI), particularly convolutional neural networks (CNNs), offers potential as a non-invasive alternative using computed tomography (CT) imaging. The aim of the study was to evaluate the performance of a CNN model in predicting urinary stone composition from non-contrast CT (NCCT) images.Material and methods A retrospective analysis was conducted using an institutional database of patients with urinary stones surgically removed between 2020 and 2024. Stone composition was confirmed via FTIR. A total of 1,487 NCCT scan images were used for CNN development and internal validation, with 105 scan images reserved for validation. Transfer learning with VGG16 architecture was applied, incorporating data augmentation and fine-tuning strategies. Performance metrics included accuracy, area under the ROC curve (AUC), and Cohen's κ coefficient.Results The dataset comprised calcium oxalate monohydrate (74.9%), calcium oxalate dihydrate (17.6%), uric acid (2.7%), struvite (2.3%), and other compositions (2.5%). The CNN achieved an internal accuracy of 89.8% (AUC 0.92). In independent temporal validation the cohort accuracy was 85.4%, with an AUC of 0.81 and a Cohen's κ of 0.864. Misclassifications occurred mainly between calcium oxalate subtypes due to overlapping radiological features.Conclusions CNN-based analysis of NCCT images provides accurate, non-invasive prediction of urinary stone composition, with performance comparable to conventional methods. This approach may improve preoperative planning and broaden diagnostic access, particularly in settings where FTIR is unavailable. Prospective multicenter studies are warranted to validate clinical implementation.
SN - 2080-4806
SN - 2080-4873
ER -