基于整整小波变换的心电数据压缩算法

根据整整小波变换理论提出了一种心电（ｅｌｅｃｔｒｏｃａｒｄｉｏｇｒａｍ） 数据压缩算法，整 整映射的整数运算保证了算法的快速性．由于屏蔽函数的引入和对阈值函数进行了特殊考虑，使该算法可同时取得高的数据压缩率和信息保真度．它不仅可完整保留ＱＲＳ、Ｐ、Ｔ、ＳＴ 段等临床信息，而且可有效保留心室晚电位信息．用ＭＩＴＢＩＨ 心律失常数据库对提出的压缩算法进行测试，结果表明平均压缩比为６－７ ，平均均方误差（ＰＲＤ） 为３ ．１ ％

Compression of Electrocardiogram Data Using Integer to Integer Wavelet Transform

Wavelet transform (WT) of time scale representation has found its application in biomedical engineering. It is particularly suitable for compressing electrocardiogram signal that contains different frequency components with time variance. Data reduction with high signal fidelity can thus be achieved by application of a shelter function with thresholds at different scales. This algorithm can retain the QRS, P, T, ST segment completely and also preserve the ventricular late potentials (VLP) information. The MIT?BIH arrhythmia database was used to test the algorithm. The results showed that the average data compression ratio is 6.7 and the average percent root mean?square difference is 3 1%.