T. Exarchos, A. Tzallas, D. Fotiadis, S. Konitsiotis, S. Giannopoulos
{"title":"A data mining based approach for the EEG transient event detection and classification","authors":"T. Exarchos, A. Tzallas, D. Fotiadis, S. Konitsiotis, S. Giannopoulos","doi":"10.1109/CBMS.2005.7","DOIUrl":null,"url":null,"abstract":"An automated methodology which detects transient events in EEG recordings and classifies those as epileptic spikes, muscle activity, eye blinking activity and sharp alpha activity is presented. It is based on data mining algorithms and includes four stages: (I) EEG preprocessing and transient events detection, (II) clustering of transient events and feature extraction, (III) feature discretization and (IV) association rule mining and classification. The methodology is evaluated using a dataset of 25 EEG recordings and the obtained overall accuracy is 84.35%. The major advantage of our approach is that it is able to provide interpretation for the decisions made since it is based on a set of association rules.","PeriodicalId":119367,"journal":{"name":"18th IEEE Symposium on Computer-Based Medical Systems (CBMS'05)","volume":"64 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2005-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"14","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"18th IEEE Symposium on Computer-Based Medical Systems (CBMS'05)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CBMS.2005.7","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 14
Abstract
An automated methodology which detects transient events in EEG recordings and classifies those as epileptic spikes, muscle activity, eye blinking activity and sharp alpha activity is presented. It is based on data mining algorithms and includes four stages: (I) EEG preprocessing and transient events detection, (II) clustering of transient events and feature extraction, (III) feature discretization and (IV) association rule mining and classification. The methodology is evaluated using a dataset of 25 EEG recordings and the obtained overall accuracy is 84.35%. The major advantage of our approach is that it is able to provide interpretation for the decisions made since it is based on a set of association rules.