Diagnostic performance of a coronary CT angiography-based deep learning model for the prediction of vessel-specific ischemia

Abstract:Objectives Fractional flow reserve (FFR) and instantaneous wave-Free Ratio (iFR) pressure measurements during invasive coronary angiography (ICA) are the gold standard for assessing vessel-specific ischemia. Artificial intelligence has emerged to compute FFR based on coronary computed tomography angiography (CCTA) images (CT-FFRAI). We assessed a CT-FFRAI deep learning model for the prediction of vessel-specific ischemia compared to invasive FFR/iFR measurements. Materials and methods We retrospectively selected 322 vessels from 275 patients at two centers who underwent CCTA and invasive FFR and/or iFR measurements during ICA within three months. A junior and senior radiologist at each center supervised vessel centerline-building to generate curvilinear reformats that were processed for CT-FFRAI binary outcomes (<= 0.80 or > 0.80) prediction. Reliability for CT-FFRAI outcomes based on radiologists' supervision was assessed with Cohen's kappa. Diagnostic values of CT-FFRAI were calculated using invasive FFR <= 0.80 (n = 224) and invasive iFR <= 0.89 (n = 238) as the gold standard. A multinomial logistic regression model, including all false-positive and false-negative cases, assessed the impact of patient- and CCTA-related factors on diagnostic values of CT-FFRAI. Results Concordance for CT-FFRAI binary outcomes was substantial (kappa = 0.725, p < 0.001). Sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy of CT-FFRAI in predicting vessel-specific ischemia on a per-vessel analysis, based on senior radiologists' evaluations, were 85% (58/68) and 91% (78/86), 82% (128/156) and 78% (119/152), 67% (58/86) and 70% (78/111), 93% (128/138) and 94% (119/127), and 83% (186/224) and 83% (197/238), respectively. Coronary calcifications significantly reduced the diagnostic accuracy of CT-FFRAI (p < 0.001; OR, 1.002; 95% CI 1.001-1.003). Conclusion CT-FFRAI demonstrates high diagnostic performance in predicting vessel-specific coronary ischemia compared to invasive FFR and iFR. Coronary calcifications negatively affect specificity, suggesting that further improvements in spatial resolution could enhance accuracy.

Publication year:2025

Keywords:Artificial intelligence, Deep learning, Coronary artery disease, Coronary computed tomography angiography, CT-derived fractional flow reserve, Radiology & nuclear medicine

Accessibility:Open

Review status:Peer-reviewed