Parameter decision of independent component analysis for low SNR magnetocardiogram signal

Abstract

Event Abstract Back to Event Parameter decision of independent component analysis for low SNR magnetocardiogram signal Koichiro Kobayashi1*, Masahito Yoshizawa1 and Yoshinori Uchikawa2 1 Iwate University, Japan 2 Tokyo Denki University, Japan Recently, clinical applications of magnetocardiograms (MCGs) have been widely studied. It is useful diagnosis such as the conduction process and evaluation of ischemic heart disease. However, it is difficult to make a correct diagnosis that MCGs are very low signal to noise ratio (SNR), for example MCGs measured outside a magnetically shielded room. Many noise reduction methods are proposed for biomagnetic measurements, especially independent component analysis (ICA) is a useful method for separating the signal and the noise. But ICA has some problems that are to decide many parameters and to distinguish signal or noise components from separated components. In many case, these were decided by experiential results. It was difficult to decide optimum parameters. We proposed the method of independent components that separated by the ICA and the decision method of the number of the contraction dimension. As a method of independent components, the components which synchronized with the autocorrelation of electrocardiogram and that of independent components are chosen as the signal components. As a dimension contraction method, the dimension of which the highest autocorrelation in the each contraction dimension is chosen. In order to evaluate the proposed method, the simulation study was carried out for various SNR MCGs. For the comparison, the method using linear discriminant analysis was carried out. In the component of MCG (SNR= -10dB), the correlation coefficient of the proposed method becomes 0.9 which is 0.2 higher than that of linear discriminant analysis. In the dimension contraction, the proposed method agreed almost with largest correlation coefficient calculated from all contraction dimensions. It was shown that proposed method was effective for the decision method of independent components and the number of the contraction dimension. Conference: Biomag 2010 - 17th International Conference on Biomagnetism , Dubrovnik, Croatia, 28 Mar - 1 Apr, 2010. Presentation Type: Poster Presentation Topic: MCG: Instrumentation, Modeling, Basic and Clinical Studies Citation: Kobayashi K, Yoshizawa M and Uchikawa Y (2010). Parameter decision of independent component analysis for low SNR magnetocardiogram signal. Front. Neurosci. Conference Abstract: Biomag 2010 - 17th International Conference on Biomagnetism . doi: 10.3389/conf.fnins.2010.06.00314 Copyright: The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers. They are made available through the Frontiers publishing platform as a service to conference organizers and presenters. The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated. Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed. For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions. Received: 06 Apr 2010; Published Online: 06 Apr 2010. * Correspondence: Koichiro Kobayashi, Iwate University, Iwate, Japan, kobaya@iwate-u.ac.jp Login Required This action requires you to be registered with Frontiers and logged in. To register or login click here. Abstract Info Abstract The Authors in Frontiers Koichiro Kobayashi Masahito Yoshizawa Yoshinori Uchikawa Google Koichiro Kobayashi Masahito Yoshizawa Yoshinori Uchikawa Google Scholar Koichiro Kobayashi Masahito Yoshizawa Yoshinori Uchikawa PubMed Koichiro Kobayashi Masahito Yoshizawa Yoshinori Uchikawa Related Article in Frontiers Google Scholar PubMed Abstract Close Back to top Javascript is disabled. Please enable Javascript in your browser settings in order to see all the content on this page.