Twistable and Stretchable Nasal Patch for Monitoring Sleep-Related Breathing Disorders Based on a Stacking Ensemble...

Area of Science:

• Biomedical Engineering
• Sleep Medicine
• Machine Learning

Background:

• Obstructive sleep apnea syndrome (OSAS) negatively impacts sleep, homeostasis, and cognitive functions.
• Current polysomnography methods are conducted in unfamiliar settings, potentially compromising sleep pattern accuracy.
• There is a need for convenient and accurate methods for diagnosing sleep disorders in a natural environment.

Purpose of the Study:

• To develop and evaluate a machine learning-based, wearable patch system for diagnosing obstructive sleep apnea syndrome (OSAS) in daily life.
• To create a non-invasive, personalized diagnostic tool that enhances patient comfort and data accuracy.
• To explore the potential of this technology as an early detection platform for chronic diseases.

Main Methods:

• A stretchable, twistable patch system was designed for direct nasal application.
• The patch simultaneously senses microscopic vibrations and airflow (≥0.1 m/s) in the nasal cavity and paranasal sinuses.
• A stacking ensemble learning model was employed to predict the degree of sleep-disordered breathing.

Main Results:

• The patch demonstrated high sensitivity and flexibility, with excellent linearity (R² = 0.992) concerning curvature.
• The machine learning model achieved a high diagnostic accuracy of 92.9% for sleep-disordered breathing.
• The system provides proactive visual notifications for detected sleep disorders.

Conclusions:

• The developed patch system offers a promising, accurate, and comfortable solution for diagnosing obstructive sleep apnea syndrome (OSAS) in everyday settings.
• This technology has the potential to serve as a valuable diagnostic platform for the early detection of associated chronic conditions like cerebrovascular disease and diabetes.
• Personalized, wearable diagnostic tools represent a significant advancement in sleep medicine and proactive healthcare.