Analysis of Respiratory Sound Detection in Lungs Using Machine Learning Techniques

Abstract

Healthy life enhances the quality of life by allowing individuals in engage able activities and build meaning full relationship for every human. Exploring to air pollution and viral infection can exacerbate or weaken the immune systems defences, making individuals more susceptible to viral respiratory infections such as influenza, asthma, Chronic Obstructive Pulmonary Diseases (COPD), Respiratory Syncytial Virus (RSV), bronchitis and pneumonia. These viral pathogens can worsen existing lung conditions and increases vulnerability leads to respiratory diseases. The virus is mutable continuously, so scientists need more time to make a medical solution for the virus. Based on this reason, high deaths are discovered every day in our life. Thus, a primary shield for protection is required to control outbreaks and pandemics. Hence it is essential that everyone should have a Small portable Lung Disease detection Device that analyses the breath sound. The proposed methodology, based on the Audio sound is used to diagnose the disease in the lungs with assistant of machine learning techniques; our aim is to increases the correct prediction rates and diminishes the false detection rate. It seems that the random forest classifier performs well on this dataset, with a training accuracy of 1.0 and a testing accuracy of 0.95.