HRV feature selection based on discriminant and redundancy analysis for neonatal seizure detection

Malarvili, M. B., Mesbah, M. and Boashash, B. (2007). HRV feature selection based on discriminant and redundancy analysis for neonatal seizure detection. In: Information, Communications and Signal Processing 2007. 6th International Conference on Information, Communications and Signal Processing 2007 (ICICS 2007), Meritus Mandarin Hotel, Singapore, (1-5). 10-13 December 2007. doi:10.1109/ICICS.2007.4449765


Author Malarvili, M. B.
Mesbah, M.
Boashash, B.
Title of paper HRV feature selection based on discriminant and redundancy analysis for neonatal seizure detection
Conference name 6th International Conference on Information, Communications and Signal Processing 2007 (ICICS 2007)
Conference location Meritus Mandarin Hotel, Singapore
Conference dates 10-13 December 2007
Convener IEEE; Nanyang Technological University
Proceedings title Information, Communications and Signal Processing 2007
Place of Publication Piscataway, NJ, U.S.A.
Publisher IEEE - Institute of Electrical Electronics Engineers Inc.
Publication Year 2007
Year available 2007
Sub-type Fully published paper
DOI 10.1109/ICICS.2007.4449765
ISBN 1-4244-0983-7
978-1-4244-0983-9
Start page 1
End page 5
Total pages 5
Language eng
Abstract/Summary This paper addresses the feature selection problem by using a discriminant and redundancy based method to select a feature subset with high discriminatory power between the classes of newborn heart rate variability (HRV) corresponding to seizure and non-seizure. The proposed method combines the Fast Correlation Based Filter (FCBF) criteria for redundancy analysis with the area under the Receiver Operating Curves (AUC) for discriminant analysis. The classification accuracies of the selected features were compared using 3 different classifiers, namely linear classifier, quadratic classifier and k-Nearest Neighbour (k- NN) statistical classifiers in a leave-one-out (LOO) cross validation. It was found that the 1-NN outperformed the other classifiers resulting in a significant reduction in feature dimensionality while achieving 85.7% sensitivity and 84.6% specificity.
Subjects 090609 Signal Processing
0903 Biomedical Engineering
Keyword Heart rate variability
Feature extraction
Feature selection-filter
Statistical classifier
Neonatal seizure detection
Q-Index Code E1

Document type: Conference Paper
Sub-type: Fully published paper
Collections: Excellence in Research Australia (ERA) - Collection
UQ Centre for Clinical Research Publications
 
Versions
Version Filter Type
Citation counts: Scopus Citation Count Cited 12 times in Scopus Article | Citations
Google Scholar Search Google Scholar
Created: Thu, 26 Mar 2009, 22:47:52 EST by Mary-Anne Marrington on behalf of UQ Centre for Clinical Research