Title: Heart murmur detection/classification using Cochlea-Like Pre-Processing and Artificial Intelligence

Authors: W. Ahmad; M.I. Hayee; J.L. Fitzakerley; S. Burns; G. Nordehn

Addresses: Department of Electrical Engineering, Upson Hall II Room 160, 243 Centennial Drive Stop 7165, University of North Dakota Grand Forks, ND 58202-7165, USA. ' Department of Electrical Engineering, 271 Marshall W. Alworth Hall, 1023 University Dr., University of Minnesota Duluth, Duluth, MN 55812, USA. ' Department of Physiology and Pharmacology, University of Minnesota Medical School Duluth, 307 Smed, 1035 University Drive, Duluth, MN 55812, USA. ' Department of Electrical Engineering, 271 Marshall W. Alworth Hall, 1023 University Dr., University of Minnesota Duluth, Duluth, MN 55812, USA. ' Department of Family and Community Health, University of Minnesota Medical School Duluth, 153 Smed, 1035 University Drive, Duluth, MN 55812, USA

Abstract: In this research paper, we used a novel approach to pre-process the heart sound signals by altering the electrical signal in a similar way as is done by human cochlea before they go to Artificial Intelligence (AI) for murmur detection/classification. Cochlea-like pre-processing changes the spectral contents of the heart sounds to enhance the murmur information which can then be detected/classified more accurately by AI circuitry. We designed a heart murmur detection/classification system based upon this approach and tested this system using simulated sounds of various murmur types. Our test results show that this approach significantly improves heart murmur detection/classification accuracy.

Keywords: heart murmurs; heart murmur detection; heart murmur classification; AI; artificial intelligence; human cochlea; CLPP; cochlea-like pre-processing; ANNs; artificial neural networks; heart sound signals.

DOI: 10.1504/IJBET.2011.042500

International Journal of Biomedical Engineering and Technology, 2011 Vol.7 No.1, pp.87 - 96

Published online: 21 Jan 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article