Ant K-Means Clustering Method on Epileptic Spike Detection

Author

Shen, Tsu-Wang ; Kuo, Xavier ; Hsin, Yue-Loong

Author_Institution

Dept. of Med. Inf., Tzu Chi Univ., Hualien, Taiwan

Volume

6

fYear

2009

fDate

14-16 Aug. 2009

Firstpage

334

Lastpage

338

Abstract

Sudden unexpected death in epilepsy (SUDEP) is the top of death rate of epilepsy population. To develop an accurate, realizable, personalized automatic epilepsy spike detection method is valuable for understanding epilepsy and preventing the possible loss. In this research, a novel spike detection method based on ant k-means (AK) clustering is proposed. By compare with other intelligent computing methods, our results show that AK worked successfully well in our epilepsy patient data with 100% sensitivity, 96% specificity, and 97.9% accuracy. Although the EEG analysis system still has room for improving, the preliminary results are encouraging for future developments.

Keywords

electroencephalography; fuzzy set theory; medical signal processing; neurophysiology; optimisation; patient diagnosis; patient treatment; pattern clustering; EEG analysis system; ant k-means clustering method; automatic epilepsy spike detection; epilepsy population; intelligent computing methods; patient data; sudden unexpected death; Ant colony optimization; Artificial intelligence; Biomedical informatics; Biomimetics; Clustering methods; Electroencephalography; Epilepsy; Feature extraction; Filters; Nervous system; Ant colony optimization; Biomimetic computing; Clustering analysis; Intelligent computing; Spike detection;

fLanguage

English

Publisher

ieee

Conference_Titel

Natural Computation, 2009. ICNC '09. Fifth International Conference on

Conference_Location

Tianjin

Print_ISBN

978-0-7695-3736-8

Type

conf

DOI

10.1109/ICNC.2009.639

Filename

5365187