Published Paper Details:

Sleep-related disorders are prevalent for older people in our modern society. Among many sleep disorders, sleep apnea is often found to be critical. A subject suffering from sleep apnea often found loud snoring during sleep due to restriction in the upper airway. As per standard, sleep apnea/hypopnea is defined as complete or partial cessation of airflow to the lungs during sleep for 10 sec or more. For treatment, assessment of apnea severity is essential and therefore requires sleep study. The laboratory-based sleep study is cost-intensive and also scare. This work attempts to develop a FUZZY based automated apnea event detection system with the help of an oro-nasal airflow signal that can be used in home-based monitoring of suspected subjects. Two-time domain-based indexes have been extracted from the measured oro-nasal airflow signal: instantaneous volume flow rate and variance. These indices are fed to the FUZZY system for the detection of Apnea/Hypopnea events. This work analyses overnight sleep recordings of oro-nasal airflow signals from 16 subjects diagnosed with obstructive sleep apnea syndrome. For validation, an independent test has been performed on four randomly chosen recordings. The result was found to be very promising, with 92% accuracy in the case of apnea event detection and 85.6% accuracy for hypopnea event detection. The proposed system can be used for home-based screening of suspected apneic subjects and can count a total number of apnea and hypopnea events during sleep.