Computer-Aided Diagnosis System for Chronic Obstructive Pulmonary Disease Using Empirical Wavelet Transform on Auscultation Sounds

Abstract In this study, it is aimed to develop computer-aided a diagnosis system for Chronic Obstructive Pulmonary Disease (COPD) which is a completely incurable and chronic disease. The COPD causes obstructions of the airways in the lungs by arising air pollution environments. Contributing analysis of abnormalities in simple ways is very important to shorten the duration of treatment by early diagnosis. The most common diagnostic method for respiratory disorders is auscultation sounds. These sounds are also essential and effective signals for diagnosing the COPD. The analysis was performed using signals from the RespiratoryDatabase@TR which consists of 12-channel lung sounds. In the computerized analysis, Empirical Wavelet Transform (EWT) algorithm was applied to the signals for extracting different modes. Afterwards the statistical features were extracted from each EWT modulation. The highest classification performances were achieved with the rates of 90.41%, 95.28%, 90.56% and 85.78% for Support Vector Machine, AdaBoost, Random Forest and J48 Decision Tree, respectively. The contribution of the study is reducing the diagnosis time to 5 seconds within higher accuracy rate.