A novel real-time multilead ECG compression and de-noising method based on the wavelet transform

A new de-noising and compression method for ECG signals has been developed based on the wavelet transform. It has been designed for mobile telecardiology scenarios, where reliability as well as spectral efficiency are essential. The signal is segmented into beats and a beat template is subtracted to them. Beat templates as well as residual signals are coded with a wavelet expansion. De-noising and compression are achieved by selecting a subset of wavelet coefficients. The number of coefficients depends on the noise level. A SNR improvement of about 5 dB was obtained. Compression performance has been tested using a subset of ECG records from MIT-BlH arrhythmia database. For example, a compression ratio of 35 with a PRD as low as 3.6% was achieved for record 119.