Summary
IEICE Information and Communication Technology Forum
2013
Session Number:ENG
Session:
Number:ENG-4
A Study of Filter-Bank Processing of ECG Signal for Diagnostic Applications
Emir Turajlic,
pp.-
Publication Date:2013-07-20
Online ISSN:2188-5079
Summary:
Processing and classification of electrocardiogram (ECG) recordings are some of the most challenging fields of biomedical signal processing owing to the fact that ECG signals commonly exhibit complex temporal morphology and contain numerous artifacts of data collection process. This paper presents study of filter bank based processing of ECG signal for the purpose of atrial fibrillation diagnostics. The examined diagnostic system relies on the statistical description of signal’s energy distribution in the individual filter banks as a feature vector for the considered classification algorithms, namely Support Vector Machines (SVM) and Artificial Neural Networks (ANN). The considered statistical measures include mean, variance, skewness and kurtosis. The effect of filter bank number on the ability to differentiate between atrial fibrillation and healthy ECG signals is examined and the diagnostic relevance of each statistical parameter is also ascertained. A systematic study of diagnostic accuracy is imposed on the choice of feature vector, whereby various combinations of filter banks and the statistical measures are evaluated. The results show that the an optimal selection of sub-bands in conjunction with the appropriate choice of statistical descriptors can lead to a considerable reduction in the feature vector size without adverse effects on classification accuracy levels.