A Compact Unified Model for Moderate-to-Severe Carotid Artery Stenosis Using Image Segmentation: A Multicenter Study
7 affiliations
- 1Department of Computer Engineering, Özyeğin University, İstanbul, Türkiye
- 2Department of Cardiology, Faculty of Medicine, Bakırçay University, İzmir, Türkiye
- 3Department of Family Medicine, University of Health Sciences, İzmir, Türkiye
- 4Department of Cardiovascular Surgery, Central Hospital, İzmir, Türkiye
- 5Department of Cardiovascular Surgery, Faculty of Medicine, University of Başkent, İzmir, Türkiye
- 6University of Health Sciences, Faculty of Medicine, Department of Cardiovascular Surgery, İzmir, Türkiye
- 7Department of Cardiology, Central Hospital, İzmir, Türkiye
Abstract
Background:: This study aimed to evaluate the diagnostic performance of a new artificial intelligence model, the compact unified model (CU-Model), for moderate-to-severe carotid artery stenosis using digital subtraction angiography (DSA) images compared to clinical reference standards.
Methods:: In this retrospective multicenter study, 156 patients with confirmed moderate-to-severe carotid artery stenosis were included. A pretrained CU-Model (trained on public non-clinical datasets for optical flow and stereo matching) was applied to raw DSA images without any fine-tuning on clinical data to avoid data leakage. The model served exclusively as a post-acquisition support tool: cardiologists applied it immediately after procedures for automated vessel segmentation and highlighting. Final North American Symptomatic Carotid Endarterectomy Trial (NASCET) measurements and interpretations were performed manually under expert supervision by 2 blinded researchers.
Results:: The CU-Model identified left internal carotid artery (ICA) stenosis in 84 patients and right ICA stenosis in 72. It misclassified only 1 moderate left ICA stenosis case as normal, achieving 99.4% sensitivity (155/156) for any stenosis detection. For stenosis severity grading, strong agreement with DSA was observed (moderate: 57.68 ± 5.94 vs. 57.71 ± 6.19, r = 0.83, P < .001, 95% CI: 0.76-0.88; severe: 86.96 ± 5.97 vs. 87.02 ± 6.18, r = 0.91, P < .001, 95% CI: 0.85-0.95). Of 156 cases, 106 were moderate (50%-69%) and 50 severe (≥70%), using NASCET-adapted thresholds accounting for measurement variability. Hypertension (77%), smoking, and syncope were common, with significant association in symptomatic patients (P < .05).
Conclusion:: The CU-Model demonstrates strong concordance with DSA as a practical, post-procedure support tool for vessel segmentation. Pending larger validation studies, it shows promise for enhancing clinical workflow efficiency.
Related
- Diagnostic Performance of a Deep Learning Model Using Phonocardiogram and Single‐Lead ECG Signals for Detecting Moderate‐or‐Greater Aortic Stenosis: Multicenter External Validation
- Study protocol for a prospective multicenter longitudinal cohort study on progression of moderate aortic valve stenosis, the MODAS Study
- Early Valve Replacement vs Surveillance in Moderate Aortic Stenosis: Meta-Analysis of Randomized and Observational Evidence
- Distinct Biomarker Signatures of CMR-Defined Myocardial Remodeling in Moderate Aortic Stenosis
- Short-term clinical and radiographic outcomes of unilateral biportal endoscopic decompressive laminectomy for lumbar spinal stenosis with mild-to-moderate vertebral rotation: a retrospective study
BibTeX
@article{tavlı2026compact,
title = {{A Compact Unified Model for Moderate-to-Severe Carotid Artery Stenosis Using Image Segmentation: A Multicenter Study}},
author = {Ahmet Tavlı and İlker Gül and Haluk Mergen and Bülent Çümen and Murat Ertürk and Haydar Yaşa and Talat Tavlı},
journal = {Anatolian Journal of Cardiology},
year = {2026},
month = oct,
doi = {10.14744/anatoljcardiol.2026.5879},
url = {https://doi.org/10.14744/anatoljcardiol.2026.5879},
note = {CC BY-NC 4.0},
}
RIS
TY - JOUR
TI - A Compact Unified Model for Moderate-to-Severe Carotid Artery Stenosis Using Image Segmentation: A Multicenter Study
AU - Ahmet Tavlı
AU - İlker Gül
AU - Haluk Mergen
AU - Bülent Çümen
AU - Murat Ertürk
AU - Haydar Yaşa
AU - Talat Tavlı
PY - 2026
DA - 2026/10/05
JO - Anatolian Journal of Cardiology
DO - 10.14744/anatoljcardiol.2026.5879
UR - https://doi.org/10.14744/anatoljcardiol.2026.5879
AB - Background:: This study aimed to evaluate the diagnostic performance of a new artificial intelligence model, the compact unified model (CU-Model), for moderate-to-severe carotid artery stenosis using digital subtraction angiography (DSA) images compared to clinical reference standards. Methods:: In this retrospective multicenter study, 156 patients with confirmed moderate-to-severe carotid artery stenosis were included. A pretrained CU-Model (trained on public non-clinical datasets for optical flow and stereo matching) was applied to raw DSA images without any fine-tuning on clinical data to avoid data leakage. The model served exclusively as a post-acquisition support tool: cardiologists applied it immediately after procedures for automated vessel segmentation and highlighting. Final North American Symptomatic Carotid Endarterectomy Trial (NASCET) measurements and interpretations were performed manually under expert supervision by 2 blinded researchers. Results:: The CU-Model identified left internal carotid artery (ICA) stenosis in 84 patients and right ICA stenosis in 72. It misclassified only 1 moderate left ICA stenosis case as normal, achieving 99.4% sensitivity (155/156) for any stenosis detection. For stenosis severity grading, strong agreement with DSA was observed (moderate: 57.68 ± 5.94 vs. 57.71 ± 6.19, r = 0.83, P < .001, 95% CI: 0.76-0.88; severe: 86.96 ± 5.97 vs. 87.02 ± 6.18, r = 0.91, P < .001, 95% CI: 0.85-0.95). Of 156 cases, 106 were moderate (50%-69%) and 50 severe (≥70%), using NASCET-adapted thresholds accounting for measurement variability. Hypertension (77%), smoking, and syncope were common, with significant association in symptomatic patients (P < .05). Conclusion:: The CU-Model demonstrates strong concordance with DSA as a practical, post-procedure support tool for vessel segmentation. Pending larger validation studies, it shows promise for enhancing clinical workflow efficiency.
SN - 2149-2263
SN - 2149-2271
ER -