Lossless and near-lossless compression of EEG signals

In this paper we study compression techniques for electroencephalograph (EEG) signals. A variety of lossless compression techniques, ranging from simple dictionary based approaches to more sophisticated context modeling techniques based on work in lossless image coding are investigated and compared. It is seen that compression ratios obtained by lossless compression are limited. Though lossy compression can yield significantly higher compression ratios while potentially preserving diagnostic accuracy, is is not usually employed due to legal concerns. Hence, we investigate near-lossless compression techniques that give quantitative bounds on the errors introduced during compression. It is observed that such techniques give significantly higher compression ratios. Simulation results with a large variety of data sets are reported.