Convolutional neural networks on risk stratification of patients with suspected coronary artery disease undergoing coronary computed tomography angiography

• Published in: The international journal of cardiovascular imaging Vol. 39; no. 6; pp. 1209 - 1216
• Main Authors: Adolf, Rafael, Nano, Nejva, Chami, Alessa, von Schacky, Claudio E., Will, Albrecht, Hendrich, Eva, Martinoff, Stefan A., Hadamitzky, Martin
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.06.2023; Springer Nature B.V
• Subjects: Algorithms; Angina; Angiography; Annotations; Artificial neural networks; Calcification; Cardiac Imaging; Cardiology; Cardiovascular disease; Cardiovascular diseases; Computed tomography; Computed Tomography Angiography; Coronary Angiography - methods; Coronary artery disease; Coronary Artery Disease - diagnostic imaging; Coronary Artery Disease - therapy; Coronary vessels; Health risks; Heart diseases; Humans; Imaging; Medical imaging; Medicine; Medicine & Public Health; Myocardial infarction; Neural networks; Neural Networks, Computer; Original Paper; Predictive Value of Tests; Prognosis; Radiology; Risk; Risk Assessment; Tomography; Tomography, X-Ray Computed - methods; Training; Vein & artery diseases; Prognosis; Convolutional neural networks; Coronary artery disease; Coronary computed tomography angiography
• ISSN: 1875-8312; 1569-5794; 1875-8312; 1573-0743
• DOI: 10.1007/s10554-023-02824-y

To assess the prognostic value of convolutional neural networks (CNN) on coronary computed tomography angiography (CCTA) in comparison to conventional computed tomography (CT) reporting and clinical risk scores. 5468 patients who underwent CCTA with suspected coronary artery disease (CAD) were included. Primary endpoint was defined as a composite of all-cause death, myocardial infarction, unstable angina or late revascularization (> 90 days after CCTA). Early revascularization was additionally included as a training endpoint for the CNN algorithm. Cardiovascular risk stratification was based on Morise score and the extent of CAD (eoCAD) as assessed on CCTA. Semiautomatic post-processing was performed for vessel delineation and annotation of calcified and non-calcified plaque areas. Using a two-step training of a DenseNet-121 CNN the entire network was trained with the training endpoint, followed by training the feature layer with the primary endpoint. During a median follow-up of 7.2 years, the primary endpoint occurred in 334 patients. CNN showed an AUC of 0.631 ± 0.015 for prediction of the combined primary endpoint, while combining it with conventional CT and clinical risk scores showed an improvement of AUC from 0.646 ± 0.014 (based on eoCAD only) to 0.680 ± 0.015 (p < 0.0001) and from 0.619 ± 0.0149 (based on Morise Score only) to 0.6812 ± 0.0145 (p < 0.0001), respectively. In a stepwise model including all prediction methods, it was found an AUC of 0.680 ± 0.0148. CNN analysis showed to improve conventional CCTA-derived and clinical risk stratification when evaluating CCTA of patients with suspected CAD.