Diagnosis of pneumonia from sounds collected using low cost cell phones

摘要: Respiratory diseases, such as pneumonia, cold, flu, and bronchitis, are still the leading causes of child mortality in world. One solution for alleviating this problem is developing affordable respiratory-health assessment methods using computerized respiratory-sound analysis. This approach has become an active research area due to recent developments electronic recording devices, stethoscopes. However, all existing require specialized equipment, which can be operated only by trained medical personals. We develop a low-cost cell phone-based rapid diagnosis method respiratory health problems. A total 367 breath sounds collected from children's hospitals order accurate models evaluation. An extensive analysis performed on sounds. Statistically significance features selected each age group ANOVA 1197 acoustic features. The model evaluated binary classification task: pneumonia vs. non-pneumonia. results showed that proposed was able effectively classify even presence environmental noises. achieved 91.98% accuracy with 92.06% sensitivity 90.68% specificity. indicate recorded mobile devices used detect effectively.