A Robust Airflow Envelope Tracking and Digitization Approach for Automatic Detection of Apnea and Hypopnea Events

Abstract

Sleep apnea hypopnea syndrome (SAHS) is a common sleep disorder that can significantly decrease the quality of life. Apnea hypopnea index, the number of apnea and hypopnea events per hour of sleep, is defined for the severity of SAHS. An automatic and accurate detection of apnea and hypopnea events can overcome the limitations of manual diagnosis of SAHS. This study explored the design of a novel automated algorithm to detect apnea and hypopnea events. From polysomnography records of the Sleep Heart Health Study, the airflow and pulse oximetry signals of 30 subjects were extracted. According to the updated American Academy of Sleep Medicine scoring manual, apnea and hypopnea events were scored by an experienced sleep physiologist. The peak signal excursion was precisely determined from the airflow envelope. An apnea event was detected by the precise determination of its pre-event baseline. A hypopnea event was detected when both the airflow reduction and oxygen desaturation were satisfied. Accordingly, the automated algorithm detected 5122 events (2215 apneas and 2907 hypopneas), against the manual scoring of 5021 events (2235 apneas and 2786 hypopneas). Strong correlations between scoring and detection of apnea, hypopnea, and combined events were achieved. The overall agreement between the scoring and detection of apnea, hypopnea, and combined events were respectively 99.1%, 95.7%, and 98.0%. This automatic algorithm is applicable to any portable sleep monitoring device for the accurate detection of apnea and hypopnea events.