Researchecg mlcardiac imagingcoronary artery diseaseinterpretability
AI-ECG Predicts Vessel-Specific Coronary Stenosis Severity
7.0
Researchers on Nov. 29, 2025 described an interpretable AI-ECG model that predicts severe or complete stenosis in four major coronary arteries referenced to coronary CT angiography (CCTA). The model achieved internal AUCs of 0.744–0.818 and external AUCs of 0.667–0.971, remained robust in a clinically normal-ECG subset, and suggests potential for scalable, noninvasive CAD screening using routine ECGs.
Key Points
- 1Reports model performance with AUCs 0.794–0.818 internal and 0.667–0.971 external across four arteries
- 2Demonstrates robustness across clinically normal-ECG subset and demographic and acquisition-time subgroups
- 3Enables vessel-level risk stratification from routine ECGs, informing scalable noninvasive CAD screening workflows
Scoring Rationale
Strong vessel-level results and external validation drive a high impact score; limited peer-review and clinical deployment constrain it.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
