Adventitious and Normal Respiratory Sound Analysis with Machine Learning Methods

The computerized respiratory sound analysis systems provide vital information concerning the current condition of lung. These systems, used by physicians for diagnosis diseases, help to classify sounds. Because each physician has different knowledge and experience, there is a problem with diagnosing treating system diseases. This study will decide in various difficult diagnostic situations easily. For this purpose, machine learning classifiers feature extraction models have been constituted sounds as healthy patient then its results were compared. In study, Empirical Mode Decomposition, Mel Frequency Cepstral Coefficients, Wavelet Transform methods are extraction, while k Nearest Neighbor, Artificial Neural Networks, Support Vector Machines classification. best accuracy was 98.8% using combination Coefficient Neighbor methods.