30 Pages

Track III: Medical instrument based system

System (CORTES) (Kortman, 1967), and analysis by synthesis ECG compressor (ASEC) (Zigel, 2000) etc. An excellent review and comparison of ECG compression methods is presented in (Jalaleddine, 1990). In transform methods, time domain ECG signal is mapped into a different domain and the transformed coefficients are compressed. A large number of transform methods have been reported, like Karhunen-Loeve Transform (KLT) (Degani, 1990), wavelet transform (Manikandan, 2014; Rajoub, 2002), Discrete Cosine Transform (DCT) (Allen, 1992) and Fast Fourier Transform (FFT) etc. Parameter extraction based ECG compression techniques extract some clinically significant features from ECG data. These features are used to reconstruct the signal. Methods belonging to this group are: peak picking, linear prediction methods, syntactic methods, neural nets methods and Long Term Prediction (LTP) (Imai, 1985).