Computer-based respiratory sound analysis: a systematic review

Abstract

Over the years, lung auscultation has been used as an effective clinical tool to monitor the state of the respiratory system. Lung auscultation provides valuable information regarding the patient's respiratory function. Recent technical advances have led to the development of computer-based respiratory sound analysis which serves as a powerful tool to diagnose abnormalities and disorders in the lung. This paper provides a comprehensive review on computer-based respiratory sound analysis techniques employed by various researchers in the past. The search for articles related to computer-based respiratory sound analysis was carried out on electronic resources such as IEEE, Springer, Elsevier, Pub Med, and ACM digital library databases. Around 55 articles were identified and were subjected to a systematic review. In this review, we examine lung sound/lung disorder, sensor used, sensor locations, number of subjects, signal processing methods, classification methods, and statistical methods employed for the analysis of lung sounds by previous researchers. A brief discussion is undertaken on the overview from the previous works. Finally, the review is concluded by discussing the possibilities and recommendations for further improvements.