Artificial Neural Network for identification of heart problem

This paper demonstrated a technique to identify heart problem based on heart sound using Artificial Neural Network (ANN). The ANN is utilized for classification of eleven types of different heart sound. The original heart sound data were collected and the data were digitally filtered, segmented using sliding window approach. Next, it is pre-processed via Fast Fourier Transform (FFT) prior to classification. Initial results proved that this technique could successfully classify the various categories of heart sound evaluated with high degree of accuracy.

[1]  T. Fitzpatrick.  Color atlas and synopsis of clinical dermatology : common and serious diseases , 1992 .

[2]  B. Tovar-Corona,et al.  Time-frequency representation of systolic murmurs using wavelets , 1998, Computers in Cardiology 1998. Vol. 25 (Cat. No.98CH36292).

[3]  M.N. Taib,et al.  Classification of heart sounds using a multilayer feed-forward neural network , 2005, 2005 Asian Conference on Sensors and the International Conference on New Techniques in Pharmaceutical and Biomedical Research.

[4]  Paul R. White,et al.  Analysing heart murmurs using time-frequency methods , 1996, Proceedings of Third International Symposium on Time-Frequency and Time-Scale Analysis (TFTS-96).

[5]  T. Olmez,et al.  Classification of heart sounds by using wavelet transform , 2002, Proceedings of the Second Joint 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society] [Engineering in Medicine and Biology.

[6]  David Kuncicky.  MATLAB Programming , 2003 .

[7]  B. B. Chaudhuri,et al.  MATLAB Programming , 2007 .

[8]  A. Tilkian,et al.  Understanding heart sounds and murmurs : with an introduction to lung sounds , 1984 .

[9]  Anil K. Jain,et al.  Artificial Neural Networks: A Tutorial , 1996, Computer.