Sleep-wake detection based on respiratory signal acquired through a pressure bed sensor

Area of Science:

• Biomedical Engineering
• Sleep Medicine
• Signal Processing

Background:

• Accurate sleep-wake staging is crucial for diagnosing sleep disorders.
• Current methods often require invasive or complex monitoring setups.
• Respiratory effort is a key physiological signal during sleep.

Purpose of the Study:

• To develop and validate an automatic sleep-wake staging algorithm using only respiratory effort.
• To assess the performance of this method in both healthy subjects and patients with sleep apnea.
• To explore the potential of contactless sensing for sleep analysis.

Main Methods:

• Respiratory effort data was collected using a Pressure Bed Sensor (PBS).
• Sleep-wake staging was performed automatically based on PBS time series.
• The algorithm was developed and evaluated using 20 all-night polysomnographies as the gold standard.
• The study included 10 healthy subjects and 10 patients with severe sleep apnea.

Main Results:

• The automatic classification achieved a sensitivity of 83.59 ± 6.79%.
• Specificity was measured at 83.60 ± 15.13%.
• Overall accuracy reached 81.91 ± 6.36% when compared to human scoring.

Conclusions:

• Respiratory effort measured by a contactless Pressure Bed Sensor is a viable method for automatic sleep-wake staging.
• This non-invasive approach demonstrates significant potential for clinical applications.
• Further research could enable the computation of key sleep indices, like sleep efficiency, using this technique.