Artificial intelligence-based identification of thin-cap fibroatheromas and clinical outcomes: the PECTUS-AI study
Authors' conclusion
AI-based OCT image analysis allows standardized identification of patients at increased risk of adverse cardiovascular outcome, offering an alternative to manual image analysis. Furthermore, AI-assisted evaluation of complete imaged segments results in better prognostic discrimatory value than evaluation of the target lesion only.
Study context
- Population
- Patients with MI undergoing optical coherence tomography (OCT) of all FFR-negative non-culprit lesions (target lesions)414 analysed
- Participants
- 438 patients
- Comparison
- AI-TCFA vs CL-TCFA
- AI-TCFAn = 143
- CL-TCFAn = 124
- Primary endpoint
- Composite of:
- All-cause death
- Non-fatal MI or unplanned revascularisation
- Excluding procedural
- Stent-related events
- Follow-up
- 2 years (±30 days)
Primary result
As reported in the abstract
Primary outcome by TCFA within target lesion · 2 years
Hazard ratio (95% CI)P value
AI-TCFA
1.99(1.02–3.90)
P = 0.04CL-TCFA
1.67(.84-3.30)
P = 0.14Further findings
Primary outcome by AI-TCFA in complete pullback
| Comparison | AI-TCFA | CL-TCFA | HR (95% CI) | P value | |
|---|---|---|---|---|---|
| AI-TCFAComplete pullback | 5.50 (1.94–15.62) | P < 0.001 | |||
| NPV | 97.6% (95% CI: 94.0%-99.3%) | Not reported in the abstract | CL-TCFA: not reported in the abstract | ||
| 0.050.21520Hazard ratio (log scale) |
Original publication
Artificial intelligence-based identification of thin-cap fibroatheromas and clinical outcomes: the PECTUS-AI study. Eur Heart J. Published 1 September 2025. doi:10.
Summarised from the published abstract; every number is checked against it