Signal Processing and Feature Engineering of Respiratory Disease

Abstract

An area of study that has recently attracted more attention is respiratory sound analysis. In fact, there is a chance that the irregularities in the early phases of a lung dysfunction may be automatically inferred in this area. In this study, we provide a technique for automatically analysing respiratory sounds. The objective is to demonstrate how well Deep learning methods work for analysing respiratory sounds. Although systems for evaluating audio signals already exist in this emerging era of new technologies, there is still a need to build a tool that can analyse audio signals, diagnose sickness at an early stage, and track the recovery of patients. By utilising Deep learning techniques, we are able to determine whether a feature vector is associated with a patient who is suffering from a lung disease.