Fast Awake OSA Screening and Characterization using Anthropometric and Sound Features
Obstructive sleep apnea (OSA) is one of the most common yet underdiagnosed sleep disorders. Undiagnosed OSA, in particular, increases the perioperative morbidity and mortality risks for OSA patients undergoing surgery requiring full anesthesia. OSA screening using the gold-standard Polysomnography (PSG) is expensive and time-consuming. This proposal presents four research projects/points to apply advanced signal processing and machine learning techniques on breathing sounds’ signals for screening OSA disorder during wakefulness. This proposal will enhance the current OSA screening algorithm during wakefulness (AWakeOSA), automate breathing phase detection, investigate the anthropometric effects on acoustic signals, predict OSA characteristics, and enhance/reduce the recording hardware setup. The main outcomes of this work will enhance the performance of the AWakeOSA algorithm as an objective, accurate, reliable and quick OSA screening tool with a high classification power during wakefulness, and providing a more robust, small size, and inexpensive OSA screening setup.