1
|
OH SHULIH, ADAM MUHAMMAD, TAN JENHONG, HAGIWARA YUKI, SUDARSHAN VIDYAK, KOH JOELENWEI, CHUA KUANGCHUA, CHUA KOKPOO, TAN RUSAN, NG EDDIEY. AUTOMATED IDENTIFICATION OF CORONARY ARTERY DISEASE FROM SHORT-TERM 12 LEAD ELECTROCARDIOGRAM SIGNALS BY USING WAVELET PACKET DECOMPOSITION AND COMMON SPATIAL PATTERN TECHNIQUES. J MECH MED BIOL 2017. [DOI: 10.1142/s0219519417400073] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/20/2022]
Abstract
The occlusion of the coronary arteries commonly known as coronary artery disease (CAD) restricts the normal blood circulation required to the heart muscles, thus results in an irreversible myocardial damage or death (myocardial infarction). Clinically, electrocardiogram (ECG) is performed as a primary diagnostic tool to capture these cardiac activities and detect the presence of CAD. However, the use of computer-aided techniques can reduce the visual burden and manual time required for the analysis of complex ECG signals in order to identify the CAD affected subjects from normal ones. Therefore, in this study, a novel computer-aided technique is proposed using 2[Formula: see text]s of 12 lead ECG signals for the identification of CAD affected patients. Each of the 2[Formula: see text]s 12 lead ECG signal beats (3791 normal and 12308 CAD ECG signal beats) are implemented with four levels of wavelet packet decomposition (WPD) to obtain various coefficients. Using the fourth-level coefficients obtained for each lead ECG signal beat, new 2[Formula: see text]s. ECG signal beats are reconstructed. Later, the reconstructed signals are split into two-fold data sets, in which one set is used for acquiring common spatial pattern (CSP) filter and the other for obtaining features vector (vice versa). The obtained features are one by one fed into k-nearest neighbors (KNN) classifier for automated classification. The proposed system yielded maximum average classification results of 99.65% accuracy, 99.64% sensitivity and 99.7% specificity using 10 features. Our proposed algorithm is highly efficient and can be used by the clinicians as an aiding system in their CAD diagnosis, thus, assisting in faster treatment and avoiding the progression of CAD condition.
Collapse
Affiliation(s)
- SHU LIH OH
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore
| | - MUHAMMAD ADAM
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore
| | - JEN HONG TAN
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore
| | - YUKI HAGIWARA
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore
| | - VIDYA K. SUDARSHAN
- Department of Biomedical Engineering, School of Science and Technology, Singapore University of Social Science, 599491, Singapore
- School of Electrical and Computer Engineering, University of Newcastle, Singapore
| | - JOEL EN WEI KOH
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore
| | - KUANG CHUA CHUA
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore
| | - KOK POO CHUA
- Department of Electronics and Computer Engineering, Ngee Ann Polytechnic, Singapore
| | - RU SAN TAN
- Department of Cardiology, National Heart Centre, Singapore
| | - EDDIE Y. K. NG
- School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore
| |
Collapse
|
2
|
Characterization of Cardiovascular Diseases Using Wavelet Packet Decomposition and Nonlinear Measures of Electrocardiogram Signal. ADVANCES IN ARTIFICIAL INTELLIGENCE: FROM THEORY TO PRACTICE 2017. [DOI: 10.1007/978-3-319-60042-0_30] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 12/25/2022]
|
3
|
Buja LM. Coronary Artery Disease: Pathological Anatomy and Pathogenesis. Coron Artery Dis 2015. [DOI: 10.1007/978-1-4471-2828-1_1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 10/23/2022]
|
4
|
Screening for coronary artery disease in respiratory patients: comparison of single- and dual-source CT in patients with a heart rate above 70 bpm. Eur Radiol 2008; 18:2108-19. [DOI: 10.1007/s00330-008-0998-6] [Citation(s) in RCA: 9] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/08/2008] [Revised: 03/03/2008] [Accepted: 03/20/2008] [Indexed: 10/22/2022]
|