Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
406451 | Neurocomputing | 2015 | 11 Pages |
In general, heart medical diagnosis devices are reliable and efficient; however, they are only present in huge or modern hospitals. Heart murmurs are one of the typical heart problems. In this paper, we propose a radial wavelet neural network (RWNN) classifier for heart murmurs (pulmonary insufficiency and tricuspid insufficiency). The extended Kalman filter (EKF) is used as a learning algorithm for the RWNN. The network inputs are dimensional features, extracted from real cardiac cycles, and three classification outputs. Proposed model classification accuracy is compared with a multilayer perceptron trained with Levenberg–Marquardt training algorithm and with extreme learning machine one. The proposed model is trained and tested using real heart cycles in order to show the applicability of the proposed scheme.