Automated Detection of Sleep Apnea and Hypopnea Events Based on Robust Airflow Envelope Tracking in the Presence of Breathing Artifacts

摘要: The paper presents a new approach to detection of apnea/hypopnea events, in the presence artifacts and breathing irregularities, from single-channel airflow record. proposed algorithm, based on robust envelope detector, identifies segments signal affected by high amplitude modulation corresponding events. It is shown that envelope—free artifacts—improves effectiveness diagnostic process allows one localize beginning end each episode more accurately. performance approach, evaluated 30 overnight polysomnographic recordings, was assessed using measures such as accuracy, sensitivity, specificity, Cohen's coefficient agreement; achieved levels were equal 95 $\%$ , 90 96 0.82, respectively. results suggest algorithm may be implemented successfully portable monitoring devices, well software-packages used sleep laboratories for automated evaluation syndrome.