부산대학교 도서관

Since epileptic seizure is unpredictable and paroxysmal, an automatic system for seizure detecting could be of greatsignificance and assistance to patients and medical staff. In this paper, a novel method is proposed for multichannel patientspecificseizure detection applying the earth mover’s distance (EMD) in scalp EEG. Firstly, the wavelet decomposition isexecuted to the original EEGs with five scales, the scale 3, 4 and 5 are selected and transformed into histograms andafterwards the distances between histograms in pairs are computed applying the earth mover’s distance as effectivefeatures. Then, the EMD features are sent to the classifier based on the Bayesian linear discriminant analysis (BLDA) forclassification, and an efficient postprocessing procedure is applied to improve the detection system precision, finally. Toevaluate the performance of the proposed method, the CHB-MIT scalp EEG database with 958 h EEG recordings from 23epileptic patients is used and a relatively satisfactory detection rate is achieved with the average sensitivity of 95.65% andfalse detection rate of 0.68/h. The good performance of this algorithm indicates the potential application for seizuremonitoring in clinical practice.