Automated CAD identification system using time–Frequency representation based on eigenvalue decomposition of ECG signals

Abstract

Coronary artery disease (CAD) is a condition where coronary arteries become narrow due to the deposition of plaque inside them. It may result in heart failure and heart attack which are life-threatening conditions. Therefore, human life can be saved by detection of CAD at an early stage. Electrocardiogram (ECG) signals can be used to detect CAD. Manual inspection of ECG recordings is not reliable as the accuracy of the outcome depends on the skills and experience of clinicians. Therefore, an automated detection method for CAD based on a time–frequency representation (TFR) known as improved eigenvalue decomposition of Hankel matrix and Hilbert transform (IEVDHM-HT) using ECG beats is proposed in the present work. Time–frequency flux (TFF), coefficient of variation (COV), and energy concentration measure (ECM) are computed from the TFR matrix and fed to the random forest classifier. The proposed method has yielded 93.77% classification accuracy.