Title :
Neural network size determination for electrocardiography inverse dipole solutions
Author :
Mizuta, Hirohisa ; Yana, Kazuo ; He, Bin ; Cohen, Richard J.
Author_Institution :
Dept. of Electron. Eng., Hosei Univ., Tokyo, Japan
Abstract :
Describes a method to find an appropriate neural network size for obtaining the inverse dipole solution for electrocardiography. A network size reduction algorithm has been proposed. Computer simulation under the single dipole model revealed that the algorithm can greatly reduce the number of hidden layer units while maintaining performance
Keywords :
electrocardiography; algorithm; computer simulation; electrocardiography inverse dipole solutions; hidden layer units; network size reduction algorithm; neural network size; single dipole model; three layer perceptron; Appropriate technology; Biomedical informatics; Cities and towns; Computer simulation; Conductors; Electrocardiography; Helium; Intelligent networks; Least squares methods; Neural networks;
Conference_Titel :
Engineering in Medicine and Biology Society, 1994. Engineering Advances: New Opportunities for Biomedical Engineers. Proceedings of the 16th Annual International Conference of the IEEE
Conference_Location :
Baltimore, MD
Print_ISBN :
0-7803-2050-6
DOI :
10.1109/IEMBS.1994.415335
