Artificial Intelligence-Led Whole Coronary Artery Optical Coherence Tomography (OCT) Analysis; Validation and Identification of Drug Efficacy and Higher-Risk Plaques
Authors' conclusion
AutoOCT-based analysis of whole coronary artery OCT identifies tissue and plaque types and measures features correlating with plaque stabilization and high-risk plaques.
Study context
- Population
- Intracoronary OCT pullbacks: model development, ex vivo autopsy arteries, IBIS-4 high-intensity statin and CLIMA studiesdevelopment: 106 patients
- Comparison
- AutoOCT vs Core laboratory
- Outcome
- Tissue and plaque classification accuracy vs histology
Methodological detailsinternal ex vivo validation vs histology, external validation vs core laboratory (IBIS-4, CLIMA), Model development: 36 212 frames, 127 whole pullbacks, 106 patients
- Internal ex vivo validation vs histology
- External validation vs core laboratory (IBIS-4, CLIMA)
- Model development: 36 212 frames, 127 whole pullbacks, 106 patients
Performance
As reported in the abstract
| Test or model | Accuracy vs histology |
|---|---|
| AutoOCT | 83% |
AutoOCT
Accuracy vs histology83%
Further findings
Plaque composition change after high-intensity statin (IBIS-4, 83 patients)
- Lesion lipid arc reduced
- 13.3° versus 12.5°
- Minimum fibrous cap thickness increased
- 18.9 µm versus 24.4 µm
- Replicated core laboratory changes
High-risk plaque features leading to patient events (CLIMA, 62 patients)
- Identified minimal lumen area <3.5 mm², lipid arc >180° and fibrous cap thickness <75 µm, similar to CLIMA core laboratory
Original publication
Artificial Intelligence-Led Whole Coronary Artery OCT Analysis; Validation and Identification of Drug Efficacy and Higher-Risk Plaques. Circ Cardiovasc Imaging. Published 25 September 2025. doi:10.
Summarised from the published abstract; every number is checked against it