Artificial intelligence evaluation of coronary computed tomography angiography for coronary stenosis classification and diagnosis

• Published in: European journal of clinical investigation Vol. 54; no. 1; pp. e14089 - n/a
• Main Authors: Lee, Dan‐Ying, Chang, Chun‐Chin, Ko, Chieh‐Fu, Lee, Yin‐Hao, Tsai, Yi‐Lin, Chou, Ruey‐Hsing, Chang, Ting‐Yung, Guo, Shu‐Mei, Huang, Po‐Hsun
• Format: Journal Article
• Language: English
• Published: England Blackwell Publishing Ltd 01-01-2024
• Subjects: Angiography; Artificial Intelligence; Cardiovascular disease; Classification; Computed tomography; Computed Tomography Angiography - methods; Constriction, Pathologic; Coronary Angiography - methods; Coronary Artery Disease; Coronary Stenosis - diagnostic imaging; Coronary vessels; Deep learning; deep learning model; Heart diseases; Humans; Machine learning; Medical imaging; Predictive Value of Tests; Radiology; Reclassification; Stenosis; Tomography; computed tomography; coronary artery disease; deep learning model; angiography; artificial intelligence
• ISSN: 0014-2972; 1365-2362
• DOI: 10.1111/eci.14089

Background
Ruling out obstructive coronary artery disease (CAD) using coronary computed tomography angiography (CCTA) is time‐consuming and challenging. This study developed a deep learning (DL) model to assist in detecting obstructive CAD on CCTA to streamline workflows.
Methods
In total, 2929 DICOM files and 7945 labels were extracted from curved planar reformatted CCTA images. A modified Inception V3 model was adopted. To validate the artificial intelligence (AI) model, two cardiologists labelled and adjudicated the classification of coronary stenosis on CCTA. The model was trained to differentiate the coronary artery into binary stenosis classifications <50% and ≥50% stenosis. Using the quantitative coronary angiography (QCA) consensus results as a reference standard, the performance of the AI model and CCTA radiology readers was compared by calculating Cohen's kappa coefficients at patient and vessel levels. The net reclassification index was used to evaluate the net benefit of the DL model.
Results
The diagnostic accuracy of the AI model was 92.3% and 88.4% at the patient and vessel levels, respectively. Compared with CCTA radiology readers, the AI model had a better agreement for binary stenosis classification at both patient and vessel levels (Cohen kappa coefficient: .79 vs. .39 and .77 vs. .40, p < .0001). The AI model also exhibited significantly improved model discrimination and reclassification (Net reclassification index = .350; Z = 4.194; p < .001).
Conclusions
The developed AI model identified obstructive CAD, and the model results correlated well with QCA results. Incorporating the model into the reporting system of CCTA may improve workflows.
In this retrospective study, 2929 DICOM files and 7945 labels were extracted from curved planar reformatted coronary computed tomography angiography (CCTA) images for CCTA CAD classification artificial intelligence (AI) model development. A modified Inception V3 deep learning‐based stenosis classification AI model could identify obstructive CAD and correlate better with the quantitative coronary angiography consensus results than CCTA radiology readers at both patient and vessel levels (Cohen kappa coefficient .79 vs. .39 and .77 vs. .40, p < .0001).