Frontiers in Physiology
Coronary artery disease (CAD) is the most common cause of death globally. Patients with suspected CAD are usually assessed by exercise electrocardiography (ECG). Subsequent tests, such as coronary angiography and coronary computed tomography angiography (CCTA) are performed to localize the stenosis and to estimate the degree of blockage. The present study describes a non-invasive methodology to identify patients with CAD based on the analysis of both rest and exercise seismocardiography (SCG). SCG is a non-invasive technology for capturing the acceleration of the chest induced by myocardial motion and vibrations. SCG signals were recorded from 185 individuals at rest and immediately after exercise. Two models were developed using the characterization of the rest and exercise SCG signals to identify individuals with CAD. The models were validated against related results from angiography. For the rest model, accuracy was 74%, and sensitivity and specificity were estimated as 75 and 72%, respectively. For the exercise model accuracy, sensitivity, and specificity were 81, 82, and 84%, respectively. The rest and exercise models presented a bootstrap-corrected area under the curve of 0.77 and 0.91, respectively. The discrimination slope was estimated 0.32 for rest model and 0.47 for the exercise model. The difference between the discrimination slopes of these two models was 0.15 (95% CI: 0.10 to 0.23, p < 0.0001). Both rest and exercise models are able to detect CAD with comparable accuracy, sensitivity, and specificity. Performance of SCG is better compared to stress-ECG and it is identical to stress-echocardiography and CCTA. SCG examination is fast, inexpensive, and may even be carried out by laypersons.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Dehkordi, Parastoo; Bauer, Erwin P.; Tavakolian, Kouhyar; Zakeri, Vahid; Blaber, Andrew P.; and Khosrow-Khavar, Farzad, "Identifying Patients With Coronary Artery Disease Using Rest and Exercise Seismocardiography" (2019). Electrical Engineering Faculty Publications. 17.