Hybrid feature vector extraction in unsupervised learning neural classifier

P. S. Kostka; E. J. Tkacz; D. Komorowski

doi:10.1109/iembs.2005.1615771

Hybrid feature vector extraction in unsupervised learning neural classifier

P. S. Kostka
, E. J. Tkacz
, D. Komorowski

Department of Medical Informatics and Artificial Intelligence (RIB/RIB1)

Medical University of Silesia in Katowice
IEEE

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

Feature extraction and selection method as a preliminary stage of heart rate variability (HRV) signals unsupervised learning neural classifier is presented. Multi-domain, mixed new feature vector is created from time, frequency and time-frequency parameters of HRV analysis. The optimal feature set for given classification task was chosen as a result of feature ranking, obtained after computing the class separability measure for every independent feature. Such prepared a new signal representation in reduced feature space is the input to neural classifier based on introduced by Grosberg Adaptive Resonance Theory (ART2) structure. Test of proposed method carried out on the base of 62 patients with coronary artery disease divided into learning and verifying set allowed to chose these features, which gave the best results. Classifier performance measures obtained for unsupervised learning ART2 neural network was comparable with these reached for multiplayer perceptron structures.

Original language	English
Title of host publication	Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	5664-5667
Number of pages	4
ISBN (Print)	0780387406, 9780780387409
DOIs	https://doi.org/10.1109/iembs.2005.1615771
Publication status	Published - 2005
Event	2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005 - Shanghai, China Duration: 1 Sept 2005 → 4 Sept 2005

Publication series

Name	Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings
Volume	7 VOLS
ISSN (Print)	0589-1019

Conference

Conference	2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005
Country/Territory	China
City	Shanghai
Period	1/09/05 → 4/09/05

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being

ASJC Scopus subject areas

Signal Processing
Biomedical Engineering
Computer Vision and Pattern Recognition
Health Informatics

Access to Document

10.1109/iembs.2005.1615771

Cite this

Kostka, P. S., Tkacz, E. J., & Komorowski, D. (2005). Hybrid feature vector extraction in unsupervised learning neural classifier. In Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005 (pp. 5664-5667). Article 1615771 (Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings; Vol. 7 VOLS). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/iembs.2005.1615771

Kostka, P. S. ; Tkacz, E. J. ; Komorowski, D. / Hybrid feature vector extraction in unsupervised learning neural classifier. Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005. Institute of Electrical and Electronics Engineers Inc., 2005. pp. 5664-5667 (Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings).

@inproceedings{9b4dfeefe2734c99adf544dd0ff375ee,

title = "Hybrid feature vector extraction in unsupervised learning neural classifier",

abstract = "Feature extraction and selection method as a preliminary stage of heart rate variability (HRV) signals unsupervised learning neural classifier is presented. Multi-domain, mixed new feature vector is created from time, frequency and time-frequency parameters of HRV analysis. The optimal feature set for given classification task was chosen as a result of feature ranking, obtained after computing the class separability measure for every independent feature. Such prepared a new signal representation in reduced feature space is the input to neural classifier based on introduced by Grosberg Adaptive Resonance Theory (ART2) structure. Test of proposed method carried out on the base of 62 patients with coronary artery disease divided into learning and verifying set allowed to chose these features, which gave the best results. Classifier performance measures obtained for unsupervised learning ART2 neural network was comparable with these reached for multiplayer perceptron structures.",

author = "Kostka, \{P. S.\} and Tkacz, \{E. J.\} and D. Komorowski",

year = "2005",

doi = "10.1109/iembs.2005.1615771",

language = "English",

isbn = "0780387406",

series = "Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "5664--5667",

booktitle = "Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005",

address = "United States",

note = "2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005 ; Conference date: 01-09-2005 Through 04-09-2005",

}

Kostka, PS, Tkacz, EJ & Komorowski, D 2005, Hybrid feature vector extraction in unsupervised learning neural classifier. in Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005., 1615771, Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings, vol. 7 VOLS, Institute of Electrical and Electronics Engineers Inc., pp. 5664-5667, 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005, Shanghai, China, 1/09/05. https://doi.org/10.1109/iembs.2005.1615771

Hybrid feature vector extraction in unsupervised learning neural classifier. / Kostka, P. S.; Tkacz, E. J.; Komorowski, D.
Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005. Institute of Electrical and Electronics Engineers Inc., 2005. p. 5664-5667 1615771 (Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings; Vol. 7 VOLS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Hybrid feature vector extraction in unsupervised learning neural classifier

AU - Kostka, P. S.

AU - Tkacz, E. J.

AU - Komorowski, D.

PY - 2005

Y1 - 2005

N2 - Feature extraction and selection method as a preliminary stage of heart rate variability (HRV) signals unsupervised learning neural classifier is presented. Multi-domain, mixed new feature vector is created from time, frequency and time-frequency parameters of HRV analysis. The optimal feature set for given classification task was chosen as a result of feature ranking, obtained after computing the class separability measure for every independent feature. Such prepared a new signal representation in reduced feature space is the input to neural classifier based on introduced by Grosberg Adaptive Resonance Theory (ART2) structure. Test of proposed method carried out on the base of 62 patients with coronary artery disease divided into learning and verifying set allowed to chose these features, which gave the best results. Classifier performance measures obtained for unsupervised learning ART2 neural network was comparable with these reached for multiplayer perceptron structures.

AB - Feature extraction and selection method as a preliminary stage of heart rate variability (HRV) signals unsupervised learning neural classifier is presented. Multi-domain, mixed new feature vector is created from time, frequency and time-frequency parameters of HRV analysis. The optimal feature set for given classification task was chosen as a result of feature ranking, obtained after computing the class separability measure for every independent feature. Such prepared a new signal representation in reduced feature space is the input to neural classifier based on introduced by Grosberg Adaptive Resonance Theory (ART2) structure. Test of proposed method carried out on the base of 62 patients with coronary artery disease divided into learning and verifying set allowed to chose these features, which gave the best results. Classifier performance measures obtained for unsupervised learning ART2 neural network was comparable with these reached for multiplayer perceptron structures.

UR - https://www.scopus.com/pages/publications/33846902317

U2 - 10.1109/iembs.2005.1615771

DO - 10.1109/iembs.2005.1615771

M3 - Conference contribution

AN - SCOPUS:33846902317

SN - 0780387406

SN - 9780780387409

T3 - Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings

SP - 5664

EP - 5667

BT - Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005

Y2 - 1 September 2005 through 4 September 2005

ER -

Kostka PS, Tkacz EJ, Komorowski D. Hybrid feature vector extraction in unsupervised learning neural classifier. In Proceedings of the 2005 27th Annual International Conference of the Engineering in Medicine and Biology Society, IEEE-EMBS 2005. Institute of Electrical and Electronics Engineers Inc. 2005. p. 5664-5667. 1615771. (Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings). doi: 10.1109/iembs.2005.1615771

Hybrid feature vector extraction in unsupervised learning neural classifier

Abstract

Publication series

Conference

UN SDGs

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this