Joint Time-Frequency Domain-Based CAD Disease Sensing System Using ECG Signals

TLDR: The proposed time-frequency matrix-based modified features are applied to detect the presence of coronary artery disease (CAD) using electrocardiogram (ECG) signals and are found to be more effective as compared to the other features.

Abstract: In this paper, the time-frequency matrix-based modified features are proposed. The proposed features are applied to detect the presence of coronary artery disease (CAD) using electrocardiogram (ECG) signals. These features are utilized to detect the presence of CAD using ECG signals. In the proposed work, ECG beats are subjected to the improved eigenvalue decomposition of Hankel matrix and Hilbert transform (IEVDHM-HT)-based method. This approach provides the time-frequency representation (TFR) of the ECG beats of both classes. Further, the time-frequency-based parameters are computed from the TFR matrix. These parameters are mixed averages time-frequency ( ${\mathrm {Avg}}_{tw}$ ), frequency average ( ${\mathrm {Avg}}_{w}$ ), and time average ( ${\mathrm {Avg}}_{t}$ ) of joint time-frequency distribution functions. In this paper, these features are extracted from the complete TFR and also for the local regions of the same TFR. These features are fed to the random tree and J48 classifiers. The proposed method has obtained an accuracy of 99.93% in the separation of CAD and normal ECG beats. The ${\mathrm {Avg}}_{w}$ features are found to be more effective as compared to the other features.