Study On Detection Method Of High Frequency Oscillation Of Epileptic Scalp EEG

Epilepsy is one of the most common nervous system diseases.There are about 9 million epileptic patients in China.Epileptic seizures lead to short-term functional disorders of the brain,which are repetitive and sudden,and bring great obstacles to the daily life of the patients,which may endanger the life of the patients.The sudden abnormal discharges of cerebral neurons in epileptic seizures are often accompanied by high frequency oscillations of EEG(High frequency oscillations,HFOs).The frequency is between 80-500 Hz and outside of the traditional range of(Electroencephalogram,EEG)measurement of electroencephalogram(0.1-70Hz).Recent studies have shown that HFOs is an important biomarker in epileptic regions.Long-term detection of HFOs in scalp EEG has important clinical value in localizing epileptic area and evaluating the effect of antiepileptic drugs in patients’ daily life.However,the sampling rate,bandwidth,accuracy,portability and cost of the existing EEG recording devices cannot meet the needs of HFOs monitoring in the family environment of epileptic patients.In order to solve the above problems,according to the characteristics of HFOs,this thesis designs and implements a set of EEG acquisition system suitable for the patient’s family environment,and validates the performance of the system in the aspects of sampling rate,precision and anti-interference through a number of experiments.All meet the system design requirements.Firstly,the analog front-end circuit of scalp EEG detection is designed,and the three-stage needle array concentric ring Laplacian dry electrode is realized.Then,by simulating the isolation technology between the front-end and the upper computer,the antinoise and electrical safety characteristics of the system are improved.The small computer Raspberry Pi is used as the main controller to realize the acquisition and control of the analog front-end,portable and wireless information transmission.The GUI interface of scalp EEG acquisition system is designed with Python language,and the data management and visualization are realized.Finally,based on the realization of HFOs measurement system,the integrity and reliability of the system were tested by electrode impedance test,Α wave measurement test,motion artifact test and median nerve stimulation-induced high frequency signal test.The experimental results show that the contact impedance of the measurement system designed in this thesis is less than 20kΩ in the frequency range of 80~1000Hz,and the noise of the system is less than 1μV,which meets the requirements of HFOs measurement.The experiments of Α wave and motion artifact show that the system can accurately measure the circadian rhythm compared with the traditional electrode measurement,and has a stronger ability to eliminate the motion interference.Electrical stimulation-induced experiments show that the system can measure the high frequency of scalp EEG with high quality.