Diagnostic Accuracy and Generalizability of a Deep Learning-Based Fully Automated Algorithm for Coronary Artery Stenosis Detection on CCTA: A Multi-Centre Registry Study

Lixue Xu; Yi He; Nan Luo; Ning Guo; Min Hong; Xibin Jia; Zhenchang Wang; Zhenghan Yang

doi:10.3389/fcvm.2021.707508

Diagnostic Accuracy and Generalizability of a Deep Learning-Based Fully Automated Algorithm for Coronary Artery Stenosis Detection on CCTA: A Multi-Centre Registry Study

Front Cardiovasc Med. 2021 Nov 5:8:707508. doi: 10.3389/fcvm.2021.707508. eCollection 2021.

Authors

Lixue Xu¹, Yi He¹, Nan Luo¹, Ning Guo², Min Hong³, Xibin Jia⁴, Zhenchang Wang¹, Zhenghan Yang¹

Affiliations

¹ Affiliated Beijing Friendship Hospital, Capital Medical University, Beijing, China.
² Shukun (Beijing) Technology Co., Ltd., Beijing, China.
³ Department of Computer Software Engineering, Soonchunhyang University, Asan-si, South Korea.
⁴ Faculty of Information Technology, Beijing University of Technology, Beijing, China.

Abstract

Aims: In this retrospective, multi-center study, we aimed to estimate the diagnostic accuracy and generalizability of an established deep learning (DL)-based fully automated algorithm in detecting coronary stenosis on coronary computed tomography angiography (CCTA). Methods and results: A total of 527 patients (33.0% female, mean age: 62.2 ± 10.2 years) with suspected coronary artery disease (CAD) who underwent CCTA and invasive coronary angiography (ICA) were enrolled from 27 hospitals from January 2016 to August 2019. Using ICA as a standard reference, the diagnostic accuracy of the DL algorithm in the detection of ≥50% stenosis was compared to that of expert readers. In the vessel-based evaluation, the DL algorithm had a higher sensitivity (65.7%) and negative predictive value (NPV) (78.8%) and a significantly higher area under the curve (AUC) (0.83, p < 0.001). In the patient-based evaluation, the DL algorithm achieved a higher sensitivity (90.0%), NPV (52.2%) and AUC (0.81). Generalizability analysis of the DL algorithm was conducted by comparing its diagnostic performance in subgroups stratified by sex, age, geographic area and CT scanner type. The AUCs of the DL algorithm in the aforementioned subgroups ranged from 0.79 to 0.86 and from 0.75 to 0.93 in the vessel-based and patient-based evaluations, both without significant group differences (p > 0.05). The DL algorithm significantly reduced post-processing time (160 [IQR:139-192] seconds), in comparison to manual work (p < 0.001). Conclusions: The DL algorithm performed no inferior to expert readers in CAD diagnosis on CCTA and had good generalizability and time efficiency.

Keywords: computed tomographic angiography; coronary artery disease; deep learning; diagnostic test; invasive coronary angiography (ICA).