간질환자 뇌파의 극파 자동검출을 윟나 매개변수선정 및 콤퓨터 알고리즘

Abstract

In this paper, we have presented a new digital computer algorithm for detection of the spike waveform appearing in the epileptic patient's electroencephalogram( EEG). The algorithm is based upon the triangular model of the EEG spikes with specific ranges of three characteristic waveform parameters (Ml, TS, TO), where M1 is the parameter indicating up and down slopes of the spikes, TS is related to the peakedness of the spikes, and TO shows the total time duration of the spikes. These three parameters were automatically computed using the mini-computer system from the EEG signals and its derivatives after low-pass digital filtering process. In the elinical test of three epileptic patients, the computer system detected the EEG spikes accurately with maximum sensitivity of 74% out of the total 19 spikes, where the three parameters were set at the condition of Ml =0.041 (v Isec), TS-=7(ms), 1'0= 20(ms).