Electrocardiogram Classification Using Wavelet Transformations

The work investigates the continuous and discrete wavelet transform to determine the corresponding features of ECG signals with variable temporal and spatial components. Discrete wavelet transform is implement as a filter bank. The approximation and refinement of wavelet coefficients from different frequency sub-bands are used to eliminate high-frequency noise, compress the signals, and there classification. Continuous wavelet transform, presented in the form of a scale diagram, is using to analyse ECG signals and develop a predictive model.