A new hybrid algorithm for ECG signal compression based on the wavelet transformation of the linearly predicted error

This paper describes a hybrid technique based on the combination of wavelet transform and linear prediction to achieve very effective electrocardiogra m (ECG) data compression. First, the ECG signal is wavelet transformed usin g four different discrete wavelet transforms (Daubechies, Coiflet, Biorthog onal and Symmlet). All the wavelet transforms are based on dyadic scales an d decompose the ECG signals into five detailed levels and one approximation . Then. the wavelet coefficients are linearly predicted, where the error co rresponding to the difference between these coefficients and the predicted ones is minimized in order to get the best predictor. In particular, the re siduals of the wavelet coefficients are uncorrelated and hence can he repre sented with fewer bits compared to the original signal. To further increase the compression rate, the residual sequence obtained after linear predicti on is coded using a newly developed coding technique. As a result, a compre ssion ratio (Cr) of 20 to 1 is achieved with percentage root-mean square di fference (PRD) less than 4%. The algorithm is compared to an alternative co mpression algorithm based on the direct use of wavelet transforms. Experime nts on selected records from the MIT-BIH arrhythmia database reveal that th e proposed method is significantly more efficient in compression. The propo sed compression scheme may find applications in digital Holter recording, i n ECG signal archiving and in ECG data transmission through communication c hannels. (C) 2001 IPEM. Published by Elsevier Science Ltd. All rights reser ved.