Sumche Man, Arie C. Maan, M. Schalij, E. E. V. D. Wall, Cees A. Swenne
{"title":"T-wave alternans ranking: Striking disagreement between two vectorcardiographic measures of repolarization heterogeneity","authors":"Sumche Man, Arie C. Maan, M. Schalij, E. E. V. D. Wall, Cees A. Swenne","doi":"10.1109/CIC.2008.4749094","DOIUrl":null,"url":null,"abstract":"In an attempt to study the importance of T-wave feature selection for T-wave alternans (TWA) analysis, we compared the alternans in two vectorcardiographic variables: maximal T-loop vector (MaxT) and the T-loop vector integral (Ti). We analyzed TWA in the 72 standard 12-lead ECGs comprised in the Physionet TWA challenge database with our research ECG/VCG processing program LEADS. We computed TWA by taking the absolute differences of the even and odd averaged beat values of Ti and MaxT (MaxT-TWA and Ti-TWA); also percentual alternans (%MaxT-TWA and %Ti-TWA) was computed. Finally, we computed both the Pearson and Kendall tau-b correlation coefficients between the MaxT-TWA and Ti-TWA, and between %MaxT-TWA and %Ti-TWA. All correlation coefficients differed significantly (P<0.01) from zero, but were relatively low (R=0.333-0.663). We conclude that T-wave features contain only in part common information; the selection of the T-wave feature in which TWA is computed deserves more attention.","PeriodicalId":194782,"journal":{"name":"2008 Computers in Cardiology","volume":"80 2","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2008 Computers in Cardiology","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/CIC.2008.4749094","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
In an attempt to study the importance of T-wave feature selection for T-wave alternans (TWA) analysis, we compared the alternans in two vectorcardiographic variables: maximal T-loop vector (MaxT) and the T-loop vector integral (Ti). We analyzed TWA in the 72 standard 12-lead ECGs comprised in the Physionet TWA challenge database with our research ECG/VCG processing program LEADS. We computed TWA by taking the absolute differences of the even and odd averaged beat values of Ti and MaxT (MaxT-TWA and Ti-TWA); also percentual alternans (%MaxT-TWA and %Ti-TWA) was computed. Finally, we computed both the Pearson and Kendall tau-b correlation coefficients between the MaxT-TWA and Ti-TWA, and between %MaxT-TWA and %Ti-TWA. All correlation coefficients differed significantly (P<0.01) from zero, but were relatively low (R=0.333-0.663). We conclude that T-wave features contain only in part common information; the selection of the T-wave feature in which TWA is computed deserves more attention.