The Diagnosis of Heart Diseases Based on PCG Signals using MFCC Coefficients and SVM Classifier

During the present paper a number of cardiovascular patients were classified using features of cepstral coefficients based on national standards. Heart sounds include a healthy heart and fourth unhealthy heart, The data used in this paper was obtained from Michigan university website. After the separation of the heart murmurs from each heart sound, the cepstral coefficients based on Mel scale for each class was calculated as feature vectors and method support vector machine was used for the classification of the diseases. The diagnostic accuracy was %100. The results indicated the accuracy, reliability, and high performance capacity in method design.