COMPRESSION OF NEUROLOGICAL SIGNALS FROM THE POINT OF VIEW OF TELEMEDICINE

Abstract

Neurological signal compression is an area of digital signal processing that can be used to convert neurological signals into an efficient encoded representation which again can be decoded to produce a close approximation. In the field of neurology, common signals of interest are electrical potentials caused by firings of millions of neurons in the human brain during various mental activities. Many applications require acquisition, storage and automatic processing of Electroencephalogram (EEG) during an extended period of time. Our main aim of this work is to compress the recorded EEG. Efficient compression techniques are desired in order to effectively store or transmit the huge amount of EEG data. In this dissertation work, three compression techniques have been implemented for the purpose of compressing the EEG. These are Discrete Cosine Transform (DCT), Wavelet Transform and Linear Predictive Coding (LPC). Out of these three, Wavelet Transform based EEG compression is a new technique. Wavelets have been successfully used in myoelectric signals compression applications, but less attention has been paid towards its application in the field of the EEG compression. The aim of this dissertation work has been centered on the implementation and comparison of EEG compression techniques. Our comparative evaluation was based on the following parameters. I. Compression Ratio (CR) 2. Compression Factor (CF) 3. Signal to noise ratio (SNR) 4. Percent Residual Difference (PRD) In this work, it is found that compression ratio in the case of LPC is not variable where as in the case of DCT and Wavelet transform based EEG compression, compression ratio is variable and quality is also good with respect to LPC. In the case of LPC expected results are not obtained and this coding is working only for few signals. It is also concluded that quality decreases by increasing the compression ratio. Quality and compression ratio is moderate in the case of DCT and Wavelet transform based EEG compression. Out of three techniques Wavelets have shown better results in all aspects