A novel way to manage and control chronic respiratory diseases

Abstract

An estimated 450 million people worldwide suffer from chronic respiratory diseases such as asthma or chronic obstructive pulmonary disease (COPD). The clinical standard of care in the diagnosis and treatment of respiratory disorders is stethoscope-based lung auscultation. Clinical signs are an integral part of the diagnosis and management of these diseases. Such use of a stethoscope, however, is limited by the episodic nature of data acquisition, as well as by the limits of human subjectivity in the recognition of symptoms. Some indications of a respiratory complication may include shortness of breath, coughing, wheezing, and labored breathing. Unfortunately, there is currently no way to objectively monitor these signs. At Strados Labs we have developed the world’s first AI-powered acoustic bio-sensor designed to bring wireless, hands-free, respiratory monitoring to clinical teams over the entire episode of care. This non-invasive clinical-grade medical device also uses proprietary machine learning algorithms to identify key changes in pulmonary sounds and breathing patterns, and to notify care teams about the respiratory health status of patients. In this way, we seek to improve care triage, reduce length of hospital stay, and avoid costly pulmonary complications. The non-invasive device captures lung sounds and chest wall motion from which it extracts key features in the time and frequency domains to identify vital respiratory symptoms. Proprietary machine learning techniques, derived from state-of-the-art speech recognition algorithms, then use the characterized data to train models that automatically label areas of interest. This process creates a closed loop system that allows the Strados device to operate autonomously and ultimately improve the management and control of chronic respiratory diseases.