The Effect of Auscultation Areas on Nonlinear Classifiers in Computerized Analysis of Chronic Obstructive Pulmonary Disease

Abstract

Today, with the rapid development of technology, various methods are being developed for computer assisted diagnostic systems. Computer assisted diagnostic systems allow an objective assessment and help physicians to diagnose. Pulmonary sounds are effective signals in the diagnosis of Chronic Obstructive Pulmonary Disease (COPD). In this study, multichannel respiratory sounds obtained by auscultation method were analyzed and the effect of auscultation points on nonlinear classifiers was investigated. In computerized analysis, EWT (Empirical Wavelet Transform) method was applied to 12-channel lung sounds which obtained from different auscultation points. Then, statistical properties based on extracted frequency components were calculated. COPD and healthy lung sounds were classified using nonlinear classifiers for each channel. At the end of the study, the effects of channels on classifiers were compared.