Wake detection capacity of actigraphy during sleep

Area of Science:

• Sleep Science and Chronobiology
• Biomedical Engineering
• Wearable Technology in Health

Background:

• Actigraphy is a non-invasive method for sleep-wake monitoring.
• Polysomnography (PSG) is the gold standard for sleep assessment.
• Accurate detection of wakefulness is crucial for evaluating sleep quality.

Purpose of the Study:

• To compare actigraphy's ability to detect wakefulness against PSG under varying sleep conditions.
• To evaluate the performance of four different actigraphy scoring algorithms in wake detection.
• To assess the impact of caffeine on actigraphy's accuracy in sleep monitoring.

Main Methods:

• Simultaneous actigraphy and PSG recordings were conducted over three nights.
• Fifteen healthy adults participated in a within-subject design.
• Sleep conditions included nocturnal sleep and two daytime recovery sleeps (placebo and caffeine).

Main Results:

• Actigraphy accuracy significantly decreased with increased wakefulness, showing low sleep specificity (<50%).
• Actigraphy consistently overestimated total sleep time and sleep efficiency, particularly in conditions with more wakefulness.
• Threshold-based algorithms were less effective at detecting wakefulness and tended to overestimate awakenings compared to regression algorithms.

Conclusions:

• Actigraphy's limited ability to detect wakefulness raises concerns about its validity for fragmented sleep.
• Its use is questionable in clinical populations with sleep disorders or in scenarios like jet lag and shift work.
• Further research is needed to improve actigraphy's reliability in complex sleep-wake scenarios.