Hybrid compression technique with data segmentation for electroencephalography data
Madyan Alsenwi, M. Saeed Darweesh, Tawfik Ismail, Hassan Mostafa, Salam Gabran · 2017
In the medical applications, a large data size of Electroencephalography (EEG) is produced due to long recording time, high sampling rate, and a large number of electrodes. Therefore, more space and bandwidth are required for efficient data transmission and storing. So, to transmit EEG data efficiently with less bandwidth and storing it in a less space, EEG data compression is a very important problem. This paper introduces an efficient algorithm for EEG compression. First, the EEG data are segmented into N segment and then transformed through Discrete Cosine Transform (DCT). The transformed coefficients are passed through a thresholding process and the values below the threshold are set to zero. Finally, the resulting coefficients are coded using the Run-Length Encoding (RLE) scheme. The EEG signal can be recovered by an inverse process. Total time for compression and reconstruction (T), Compression Ratio (CR) and Percentage Root Mean Error Difference (PRD) are evaluated in order to check the effectiveness of the proposed algorithm. Simulation results show that a significant improvement in the compression time by using data segmentation.