Digital Steth Based Pneumonia Detection (P-0119) (MFN 5486)

Abstract

Respiratory sound signal analysis helps in distinguishing normal respiratory sounds from abnormal respiratory sounds and this can be used to accurately diseases as is done by a medical specialist via auscultation. A system capable of analyzing respiratory sound can be very helpful in detection of pneumonia, asthma, tuberculosis, and heart murmur sounds as the respiratory sound signal carries information about the underlying physiology of the lungs and is used to detect presence of adventitious lung sounds which are an indication of disease .This project attempts to diagnose pneumonia, by using automated analysis of lung sounds. Such an approach minimizes the subjectivity of diagnosis inherent to current practices by physicians. Breath sounds are recorded by a physiological microphone and a stethoscope, and then analyzed in software using an algorithm. The diagnosis is based on three techniques, first we get information about patient and score the questions answered by the subject, and in second technique get the lung sound, extract some specific features from lung sound and after feature extraction make the decision based on the results of classification algorithm. At the same time the system applies an algorithm on the breath sounds to calculate breaths per minute . After executing this process make the decision based on the results of these steps.