Design And Research Of Portable Epilepsy Seizure Monitoring System

Epilepsy seizure monitoring is of great significance for guiding the treatment of epilepsy and reducing the harm for patients. Electroencephalogram is often used for epilepsy seizure monitoring. Electrodes, communication mode, portability and antiinterference ability of traditional EEG acquisition system can not meet the requirements of epilepsy monitoring. In recent years, the research of eplilepsy seizure monitoring has attracted extensive attention and lots of epilepsy monitoring systems came into being. However, some of these systems are not based on the epilepsy seizure mechanism. Some of systems only use the simple spectrum analysis. It is insufficient for them to analyse EEG. Therefore, it is of great significance to propose some new methods of epilepsy EEG monitoring and design a portable, reliable and practical epilepsy seizure monitoring system.A portable epilepsy seizure monitoring system based on embedded EEG acquisition system and Android smart operating system is proposed in this paper. More details are as follows:The pre-processing methods of epilepsy EEG are selected, analysed and simulated. These pre-processing methods include lowpass filter, 50 Hz notch filter and highpass filter. Digital denoising methods for real-time epilepsy EEG monitoring are proposed. Besides, this paper also presents the multi-channel multi-scale entropy of recurrence quantification combining two multi-scale decomposition and multi-channel analysis, on the basis of the existing entropy of recurrence quantification. Data extraction, process of multi-scale decomposition and process of method optimization are illustrated in detail. Noise of different intensity is put in neural mass model data and these methods are evaluated with these data. These methods are also used to analyse real epilepsy EEG. This paper uses statistical methods to verify their analysis capability and uses linear discriminant analysis methods to analyse their classification accuracy. Through analysis, multi-scale decomposition method and multi-channel analysis method can improve the anti-noise ability, analysis capability and classification accuracy. This paper introduces the design of EEG acquisition hardware circuit. It includes the design and implementation of active electrode, analog amplifier circuit, analog filter circuit, A/D conversion circuit, data transmission circuit and power circuit. It also introduces the upper software running on the Android smart operating system. Some key parts including the permissions, communication protocol, data transmission, display, data storage, etc are detailed and discussed in this paper. The realization of key algorithms are also illustrated in detail.