AI coronary CT features correlate with flow reserve in coronary artery disease patients

This retrospective cohort study evaluated 251 patients (753 vessels) with suspected or known coronary artery disease. The investigation examined the association between AI-derived coronary CT angiography (CCTA) features and coronary flow reserve (CFR) measured by CZT-SPECT and CT-derived fractional flow reserve (FFR-CT).

The study did not report specific numerical results for the correlations between AI features and CFR or FFR-CT. No primary outcome effect sizes, p-values, or confidence intervals were provided in the input data.

Safety and tolerability were not reported; adverse events, serious adverse events, and discontinuations were not reported.

Key limitations include the retrospective design, which limits causal inference, and the absence of reported numerical outcomes, which precludes a detailed assessment of effect magnitude. The study population was limited to patients with suspected or known coronary artery disease.

In practice, these findings suggest a potential role for AI-derived CCTA features in assessing coronary physiology, but the evidence is preliminary and requires prospective validation before clinical adoption.