Skip to Main Content
Notice of Retraction
After careful and considered review of the content of this paper by a duly constituted expert committee, this paper has been found to be in violation of IEEE's Publication Principles.
We hereby retract the content of this paper. Reasonable effort should be made to remove all past references to this paper.
The presenting author of this paper has the option to appeal this decision by contacting TPII@ieee.org.
In order to discriminate normal and abnormal heart sounds (HSs) accurately and effectively, a new method for clinical diagnosis of the heart valve diseases is proposed. The method is composed of three stages. The first stage is the preprocessing stage. During the pre-processing stage, the improved wavelet threshold shrinkage denoising algorithm is used for the noise reduction of the measured HSs. In the feature extraction stage, the normalized average Shannon energy theorem and wavelet transform are used to extract the time-frequency feature of the HSs. In the classification stage, the proposed Antibody Memory Clone Clustering Algorithm (AMCCA) based on Supervised Gath-Geva algorithm is used. To test the correct classification rate of the proposed method, 110 data (60 normal, 50 abnormal) of the aortic heart valve and 120 data (60 normal, 60 abnormal,) of the mitral heart valve are collected and analyzed, the accuracy performances are achieved by 96.2%, 100%, 96% and 96.4% respectively. Furthermore, Sammon mapping algorithm is used to project the four-dimensional feature data of HSs into a lower two-dimensional data to achieve the visualization of the classification results. The experimental results indicate that the proposed method achieves high classification accuracy, and has strong clinical application value.