Multilead ECG data compression using SVD in multiresolution domain

• We propose a multiscale SVD based compression algorithm for multilead ECG data.
• Algorithm exploits intra-beat and inter-lead correlations of wavelet coefficients.
• A new thresholding technique is proposed.
• Algorithm outperforms in terms of CR and computational time.
• Simulations are done for WBSN applications by comparing with state-of-art methods.

In this paper, multilead electrocardiogram (MECG) data compression using singular value decomposition in multiresolution domain is proposed. It ensures a high compression ratio by exploiting both intra-beat and inter-lead correlations. A new thresholding technique based on multiscale root fractional energy contribution is proposed. It selects the singular values depending on the clinical importance of the wavelet subbands. The proposed method is evaluated with the PTB Diagnostic ECG database. This compression method is embedded with a pulse amplitude modulated direct sequence-ultra wideband technology for transmission of the MECG data. This may be useful in telemonitoring services for the wireless body sensor network. A comparative study of computational time complexity has also been carried out. The results show that the proposed method can be executed at least three times faster than the existing methods. The storage efficiency is enhanced by 19 times using this method.

Figure optionsDownload as PowerPoint slide