PECTUS-AI study published in European Heart Journal

In the PECTUS-AI study, led by Rick Volleberg and Thijs Luttikholt, we performed a secondary analysis of the prospective observational PECTUS-obs trial, and investigated whether artificial intelligence (AI) can be used to identify thin-cap fibroatheromas (TCFA)—a high-risk plaque feature linked to adverse outcomes. Traditionally, TCFA detection with OCT requires expert readers and is time-consuming, limiting its clinical applicability.

In this study, OCT images from 413 patients with myocardial infarction were analyzed using both an independent core laboratory (CL-TCFA) and OCT-AID, our validated AI segmentation algorithm (AI-TCFA). At two years follow-up, AI-TCFA in target lesions was significantly associated with the composite primary outcome of death, non-fatal myocardial infarction, or unplanned revascularization. Importantly, AI-based assessment of the complete OCT pullback showed an even stronger prognostic value, with a negative predictive value of nearly 98%.

Key findings:

• AI-TCFA was identified in 34.5% of patients, CL-TCFA in 30.0%.

• AI-TCFA in target lesions predicted adverse outcomes [HR 1.99, 95% CI 1.02–3.90].

• AI-TCFA across complete pullbacks provided even higher risk discrimination [HR 5.50, 95% CI 1.94–15.62].

Conclusion:

AI-based OCT image analysis enables standardized, efficient detection of high-risk coronary plaques and provides superior prognostic value compared to conventional expert assessment. These findings support the potential of AI to transform intracoronary imaging into a powerful tool for risk stratification in daily clinical practice.

The impact of this research was widely recognized, with coverage in national media, including RTL Nieuws, RTL Tonight, De Gelderlander, and BNR Nieuwsradio.

Read the full article here!

← Back to overview