Pre-Processing of multi-channel EEG for improved compression performance using SPIHT

Abstract

A novel technique for Electroencephalogram (EEG) compression is proposed in this article. This technique makes use of the inter-channel redundancy present between different EEG channels of the same recording and the intra-channel redundancy between the different samples of a specific channel. It uses Discrete Wavelet Transform (DWT) and Set partitioning in hierarchical trees (SPIHT) in 2-D to code the EEG channels. Smoothness transforms are added in order to guarantee good performance of SPIHT in 2-D. Experimental results show that this technique is able to provide low distortion values for high compression ratios (CRs). In addition, performance results of this method do not vary a lot between different patients which proves the stability of the method when used with recordings of different characteristics.

An evaluation of the effects of wavelet coefficient quantisation in transform based EEG compression

“…1. It should be noted that an alternative definition of PRD, employed by [24], [28], [30], does not remove the signal mean prior to calculation. This inclusion of the mean (DC bias) can create an artificially low PRD, whereby the mean of the signal is maintained, while important signal information is lost.…”

“…Daou and Labeau present a 2-D SPIHT based EEG compression methodology in [30]. Improved compression performance in comparison to 1-D SPIHT and a number of other algorithms were reported for PRDs lower than 30%.…”

“…In this work they examine lossy compression approaches based on wavelet and wavelet packet transforms using the MIT-BIH Polysomnographic Database [27]. In [30], Daou and Labeau present a 2-D SPIHT based EEG compression methodology using EEG data obtained from Montreal Neurological Institute. They propose a pre-processing technique to exploit the correlation between EEG channels to maximise compression and employs two transform operations: a DWT and Discrete Cosine Transform (DCT).…”

Energy-aware cross-layer optimization for EEG-based wireless monitoring applications

“…Consequently, the unknowns in this problem are the transmission rates r i , compression ratios C Ri and wavelet filter lengths F i . By knowing the transmission rates, the required transmission energy from different nodes can then be obtained from (6). Similarly, by knowing the compression ratios and wavelet filter lengths, the required encoding energy can then be obtained from (16).…”

“…They also ignore the cross-layer design which optimizes the performance by jointly considering multiple protocol layers. In the past few years, much of the research in the area of BASNs has focused on issues related to wireless sensor designs, sensor miniaturization, signal compression techniques and low-power hardware design [3] [4][5] [6]. A good review of state-of-the-art hardware, technologies, and standards for BASN was presented in [7].…”