A new ECG data compression method based on adaptive vector quantization and residual error compensation

Abstract

In the last few years, data compression has become essential for both storage and transmission especially in the biomedical domain. In this paper, we propose a new technique for the electrocardiogram data compression. It is based on the singular value decomposition and a new adaptive vector quantization. In the proposed method, the codebook is generated adaptively at each quantization stage. The result of this procedure is a reduction of the codebook size. A residual encoding and a compensation of the residual reconstruction error technique are also proposed. Consequently, low reconstruction error was obtained. The technique was tested using arrhythmia database of the Massachusetts Institute of Technology-Beth Israel Hospital and compared to several existing methods. The obtained results were very satisfactory, since a compression ratio of 102.54 was reached for a low reconstruction error.