Cough Sound Analysis for Pneumonia and Asthma Classification in Pediatric Population

摘要: Pneumonia and asthma are the common diseases in pediatric population. The share some similarities of symptoms that make them difficult to separate without proper diagnostic tools. majority pneumonia cases occur third world countries wherein even basic tools (e.g.: x-ray) extremely rare. In these countries, WHO recommends using rapid breathing chest in-drawing as approach diagnose children with cough. As results, many patients were misdiagnosed prescribed for unnecessary antibiotic treatment. this study, we propose a cough sound analysis based method differentiate from asthma. Cough is major symptom Past studies showed acoustic sounds may carry important information related diseases. However, there no attempts use Our extracted features such Mel-frequency cepstral coefficients, non-Gaussianity score Shannon entropy. then used develop artificial neural network classifiers. Tested leave one out validation technique eighteen subjects, our achieved sensitivity, specificity Kappa 89%, 100%, 0.89 respectively. results show potential be developed tool remote areas.