Deep Learning for Coronary Stenosis Detection in Heavily Calcified Plaques at Coronary CT Angiography: A Stepwise, Multicenter Study.
Authors
Affiliations (10)
Affiliations (10)
- Department of Radiology, Beijing Anzhen Hospital, Capital Medical University, No. 2 Anzhen Rd, Chaoyang District, Beijing, PR China 100029.
- Deepwise Artificial Intelligence (AI) Lab, Deepwise Inc, Beijing, PR China.
- Department of Radiology, General Hospital of Northern Theater Command, Liaoning, PR China.
- Division of Cardiovascular Imaging, Department of Radiology and Radiological Science and Division of Cardiology, Department of Medicine, Medical University of South Carolina, Charleston, SC.
- Department of Cardiology, Beijing Daxing Distract Hospital of Integrated Chinese and Western Medicine, Beijing, PR China.
- Department of Cardiology, Beijing Anzhen Hospital, Capital Medical University, Beijing, PR China.
- Department of Computer Science, The University of Hong Kong, Hong Kong, China.
- Diagnostic and Interventional Radiology, Shanghai Jiao Tong University Affiliated First People's Hospital, Shanghai, P.R China.
- Department of Radiology, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Shandong, PR China.
- Department of Diagnostic Radiology, Jinling Hospital, the First School of Clinical Medicine, Southern Medical University, Jiangsu, PR China.
Abstract
Purpose To develop and validate a deep-learning (DL) model for automated assessment of coronary stenosis in vessels with heavily calcified plaques on coronary CT angiography (CCTA), using quantitative coronary angiography (QCA) as the reference standard. Materials and Methods A total of 10,101 CCTAs (June 2017-December 2020) from three tertiary hospitals in China were retrospectively collected for DL model development. External testing dataset 1 included 442 CCTAs (Agatston score > 300) from two independent hospitals (January 2021-May 2022) for performance evaluation. A separate external testing dataset 2 of 120 CCTAs was used for a reader study assessing whether DL assistance improved diagnostic accuracy among junior, attending, and senior radiologists. External testing dataset 3 included 150 prospectively collected CCTAs (June-July 2023) were analyzed to compare model performance against clinical reports, simulating real-world deployment. Model diagnostic performance was assessed using receiver operating characteristic (ROC) analysis, with QCA as reference. Results In external testing dataset 1, specificities for detecting ≥ 50% stenosis were 78%, 74%, 48% and AUCs were 0.89, 0.90, 0.87 at segment, vessel, and patient levels, respectively. In external testing dataset 2, DL assistance improved radiologist specificity by 7-11% (<i>P</i> < .001) with improving AUC, and increased interreader agreement (Δκ = 0.155-0.228, <i>P</i> < .05). In external testing dataset 3, the model demonstrated 53% specificity and higher AUC versus clinical reports (0.91 vs 0.76, <i>P</i> < .001). Conclusion The proposed DL model accurately detected coronary stenosis of heavily calcified plaques on CCTA and improved diagnostic performance of radiologists. © The Author(s) 2025. Published by the Radiological Society of North America under a CC BY 4.0 license.