Performance Evaluation of Neural Network and Linear Predictors for Near-Lossless Compression of EEG Signals

Abstract

This paper presents a comparison of the performances of neural network and linear predictors for near-lossless compression of EEG signals. Three neural network predictors, namely, single-layer perceptron (SLP), multilayer perceptron (MLP), and Elman network (EN), and two linear predictors, namely, autoregressive model (AR) and finite-impulse response filter (FIR) are used. For all the predictors, uniform quantization is applied on the residue signals obtained as the difference between the original and the predicted values. The maximum allowable reconstruction error delta is varied to determine the theoretical bound delta 0 for near-lossless compression and the corresponding bit rate rp. It is shown that among all the predictors, the SLP yields the best results in achieving the lowest values for delta 0 and rp. The corresponding values of the fidelity parameters, namely, percent of root-mean-square difference, peak SNR and cross correlation are also determined. A compression efficiency of 82.8% is achieved using the SLP with a near-lossless bound delta 0 = 3, with the diagnostic quality of the reconstructed EEG signal preserved. Thus, the proposed near-lossless scheme facilitates transmission of real time as well as offline EEG signals over network to remote interpretation center economically with less bandwidth utilization compared to other known lossless and near-lossless schemes.

Near-Lossless Multichannel EEG Compression Based on Matrix and Tensor Decompositions

“…Single-channel EEG compression is widely studied, and can be categorized under lossless, near-lossless, and lossy methods (see [8] and references therein). Predictive-based coders are competitive in lossless [9] and near-lossless [10,11] scenarios, but they do not support progressive transmission and hence they are of little use in practical scenarios. Combining progressive transmission and guaranteed maximum distortion (in L ∞ sense) will be crucial in real-time transmission and clinical settings.…”

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

“…Sriraam et al present a number of recent papers on EEG compression using neural networks [24], [25]. In both papers, the authors make use of predictors as part of an approach to give near-lossless compression of EEG data.…”

Multichannel EEG Compression: Wavelet-Based Image and Volumetric Coding Approach

“…Many excellent compression techniques for single-channel EEG compression have been reported so far, which can be categorized under lossless [2]- [5], near-lossless [6,7] and lossy methods [8]- [13]. Prediction-based coders are very competitive in lossless [4] and near-lossless scenarios [6,7], when the δ is small (typically 1 or 2).…”

“…Prediction-based coders are very competitive in lossless [4] and near-lossless scenarios [6,7], when the δ is small (typically 1 or 2). However, none of the aforementioned predictive coding techniques supports progressive transmission.…”