Based on the relationship between coronary artery blockages and heart sound signals, a novel processing method on heart sound signal of early diagnosis of coronary heart disease was proposed. With the wavelet analysis, the heart sound signal was decomposed and reconstructed, and the coefficient of each layer was extracted. The information content of the first and the second heart sound signal (S1, S2) was calculated from Shannon entropy. The time threshold was applied to obtain the interval between S1 and S2. All the characteristic values were combined into a matrix containing nine elements, which was regarded as the input of a BP neural network for the identification of heart sound signal. The results show that the proposed algorithm is highly accurate for the early diagnosis of coronary heart disease. The recognition rate of the simple aortic regurgitation, the aortic regurgitation, the mitral valve stenosis and mitral valve insufficiency were 73.33, 80.00, 86.67 and 93.33 % respectively. It provides a non-invasive early diagnosis method of coronary heart disease. © 2013 Springer Science+Business Media New York.
CITATION STYLE
Liu, S., Chen, G., & Chen, G. (2013). The application of wavelet analysis and BP neural network for the early diagnosis of coronary heart disease. In Lecture Notes in Electrical Engineering (Vol. 236 LNEE, pp. 165–172). Springer Verlag. https://doi.org/10.1007/978-1-4614-7010-6_19
Mendeley helps you to discover research relevant for your work.