Spektral Özellikler ve MFCC Tabanlı Özellikleri Kullanan Klasik Makine Öğrenmesi Metotlarıyla PCG Parça Sınıflandırması

Cardiovascular diseases are some of the most common today. Congenital abnormalities, caused by impaired heart rhythm, vascular occlusion, post-operation arrhythmias, attacks and irregularities in valves various cardiovascular diseases. Early recognition them is very important for obtaining positive results treatment. For this purpose, it tried to diagnose detect listening sounds coming from heart. During rhythmic work heart, contraction relaxation chambers filling discharge blood into veins create that identified with Among characteristic there can be similar rustling which indicators pathological conditions. These unexpected sounds, rustling, called murmurs. Phonocardiograph device used record these mechanical via microphone. Heart recordings captured a phonocardiograph phonocardiograms (PCGs). Expert physicians try murmurs examining PCGs. Ambient noise, squeak microphone, patient's breathing factors make task more difficult challenging. Computer-aided systems supported machine learning, signal processing artificial intelligence algorithms offer solutions help regard. In study, detection murmur PCG frames was examined. equal length, obtained fragmenting 1-second-long frames, were classified widely learning methods namely C4.5 decision tree, Naive Bayes, Support Vector Machines k-nearest neighbor. To train those classifiers we spectral features signals, averages MFCC values refined deep model inputted values. At end manuscript accuracies compared.