Heart Rhythm Analyzed via Shapelets Distinguishes Sleep From Awake

Area of Science:

• Biomedical Engineering
• Sleep Science
• Cardiology

Background:

• Accurate sleep detection is crucial for medical and practical applications like traffic safety.
• Cardio-respiratory coupling, particularly respiratory sinus arrhythmia (RSA), differs significantly between sleep and wakefulness due to vagal activity.
• Increased cardio-respiratory coupling is a key indicator of sleep onset.

Purpose of the Study:

• To develop a method for detecting sleep-wake transitions using only heart rate variability (HRV) data.
• To leverage the relationship between heart rate dynamics and respiration during different states of consciousness.

Main Methods:

• Utilized heart rate variability (HRV) data from 75 healthy individuals recorded with high precision.
• Developed a novel method based on quantifying the similarity of 'shapelets' (short HRV time series segments) related to the respiration cycle.
• Analyzed patterns indicative of cardio-respiratory coupling changes associated with sleep onset.

Main Results:

• Identified distinctive HRV patterns that are stable across age and sex.
• The method accurately distinguishes between sleep and awake states.
• The approach can pinpoint the transition from wakefulness to sleep almost immediately.

Conclusions:

• Heart rate variability analysis offers a viable, non-invasive method for detecting sleep onset.
• The developed shapelet-based method shows promise for real-time sleep detection and potentially sleep prediction.
• Further research could refine this technique for reliable sleep prediction systems.