ECG signal analysis is very essential for the diagnosis of most of the cardiac diseases ECG is a recording method of electrical impulses which are generated in the heart. The useful information about the functionality of the human heart is provided by the ECG interpretation. While the ECG is a nonlinear or non- stationary signal, the slight changes in its amplitude and duration are not well explained in time and frequency domains. The intervals and amplitudes of the ECG waves describe the different features required for the ECG signal analysis such as statistical feature, morphological feature, and temporal features etc. The P-QRS-T waves in ECG signal represent one cardiac cycle and the normal heartbeat ranges between 60 to 100 beats per minute. The signal processing techniques are an obvious choice to extract the valuable information by using ECG signal for real-time analysis. Whereas, traditional techniques for signal processing are unable to deal with the non- stationary nature of the bio-signals. Further, these extracted features are applied to the classifiers for classification in different categories of cardiac disease In this proposed paper different techniques are discussed which are proposed earlier for extracting useful features for the analysis of an arrhythmia and interpretation of PCG signals over classical processing technique with different classifiers This paper also provides a comparison of various methods proposed earlier for classification and feature extraction.
展开▼
