Automatic detection of sleep macrostructure based on a sensorized T-shirt.

Area of Science:

• Biomedical Engineering
• Sleep Science
• Wearable Technology

Background:

• Sleep evaluation traditionally relies on polysomnography (PSG).
• Wearable sensors offer a non-invasive alternative for physiological monitoring.
• Previous work validated T-shirt sensor reliability and developed classification algorithms on ECG data.

Purpose of the Study:

• To develop and apply a fully automatic procedure for sleep evaluation using a sensorized T-shirt.
• To combine T-shirt recordings with automatic classification for reliable sleep profiling.
• To classify sleep stages (WAKE, REM, NREM) using physiological signals.

Main Methods:

• Utilizing signals from a sensorized T-shirt worn during sleep.
• Applying automatic algorithms for feature extraction from physiological data.
• Classifying sleep stages based on heart rate variability (HRV), respiration, and movement signals.

Main Results:

• Demonstrated the feasibility of automatic sleep stage classification using T-shirt data.
• Achieved reliable sleep profiles, including WAKE, REM, and NREM stages.
• Validated the system's performance by comparing it to standard clinical polysomnography (PSG).

Conclusions:

• A fully automatic system for sleep evaluation using sensorized T-shirt data is effective.
• The system provides reliable sleep classification comparable to clinical standards.
• This approach offers a promising non-invasive method for at-home sleep monitoring.