Title :
A comparison of neural network models for wheeze detection
Author :
Forkheim, Kevin E. ; Scuse, David ; Pasterkamp, Hans
Author_Institution :
Dept. of Comput. Sci., Manitoba Univ., Winnipeg, Man., Canada
Abstract :
An analysis of the use of neural networks to process lung sounds and identify wheezes is presented. Both raw signal data and Fourier transform data were used to train and test a series of neural networks. The purpose of this study was to compare the performance of the neural networks and their ability to detect wheezes in isolated lung sound segments
Keywords :
Fourier transforms; backpropagation; bioacoustics; feedforward neural nets; lung; medical signal processing; patient diagnosis; pneumodynamics; self-organising feature maps; vector quantisation; Fourier transform data; asthma; backpropagation network; isolated lung sound segments; learning vector quantization network; lung sounds; neural network models; performance; radial basis function network; raw signal data; self-organising map network; wheeze detection; Acoustic noise; Computer science; Diseases; Fourier transforms; Frequency; Lungs; Microphones; Neural networks; Performance analysis; Testing;
Conference_Titel :
WESCANEX 95. Communications, Power, and Computing. Conference Proceedings., IEEE
Conference_Location :
Winnipeg, Man.
Print_ISBN :
0-7803-2725-X
DOI :
10.1109/WESCAN.1995.493973
