Development of a Software Application to Extract the Features of Normal Respiratory Sounds from the Lungs and the Trachea

Auscultation has been widely regarded as one of the most important noninvasive diagnostic tools for clinical diagnosis of the respiratory tract. The purpose of this thesis was to develop a software application capable of extracting the key features of respiratory sound signals from the lungs and trachea of healthy persons. The efficacy of the program was evaluated by the verification of the important features of the sound signals from the left and right lungs and the trachea such as 1) right and left lung symmetry and 2) dissimilarity between the trachea and both lungs. The program was developed in LabView and was designed to capture the respiratory sound signals from the lungs and the trachea in real-time and process them in the time and frequency domains for further analysis. The features compared were 1) signal amplitude in the time domain and 2) power spectra in the frequency domain. Results of the study had shown that the program had been able to verify that 1) the key features of the breath sound signals from the left and right lungs were similar and 2) the features of the signals from the trachea and both lungs were different