Research output

Performance of Dower's inverse transform and Frank lead system for Identification of Myocardial Infarction

Research output: Contribution to journalArticleAcademicpeer-review

Associated researcher

Associated organisations

Abstract

This study performs a comparison between Dower's inverse transform and Frank lead system for Myocardial Infarction (MI) identification. We have selected a set of relevant features for MI detection from the vectorcardiogram and used the lasso method after that to build a model for the Dower's inverse transform and one for the Frank leads system. Then we analyzed the performance between both models on MI detection. The proposed methods have been tested using PhysioNet PTB database that contains 550 records from which 368 are MIs. Two main conclusions are coming from this study. The first one is that Dower's inverse transform performs equally well than Frank leads in identification of MI patients. The second one is that lead positions have a large influence on the accuracy of MI patient identification.

View graph of relations

Details

Original languageEnglish
Pages (from-to)4495-4498
Number of pages4
JournalIEEE Engineering in Medicine and Biology Society
Volume2015
DOIs
Publication statusPublished - Aug 2015
EventAnnual International Conference of the IEEE Engineering in Medicine and Biology Society -
Duration: 1 Jan 20101 Jan 2010