A Wearable Stethoscope for Accurately Monitoring Real-time Lung Sounds and Distinguishing Wheezing Sounds Based on an Ai Algorithm

Abstract

The various bioacoustics signals obtained with auscultation contain complex clinical information used as traditional biomarkers, however it is not widely used in clinical for long-term studies due to spatiotemporal limitations. Here, we developed a wearable stethoscope for skin-attachable, continuous and real-time auscultation using a lung sound monitoring patch (LSMP). The LSMP can monitor respiratory function through mobile app and classify normal and adventitious breathing by comparing the unique acoustic characteristics they produced. Heart and breathing sounds from humans can be distinguished from complex sound consisting of a mixture of the bioacoustic signal and external noise. The performance was further demonstrated with pediatric asthma and elderly chronic obstructive pulmonary disease (COPD) patients. We implemented a counting algorithm to identify wheezing events in real-time regardless of the respiratory cycle. As a result, the AI-based adventitious breathing event counter distinguished over 80% of events, especially wheezing events, in long-term clinical application.