Sleep condition detection and assessment with optical fiber interferometer based on machine learning.

Area of Science:

• Biomedical Engineering
• Sensor Technology
• Sleep Medicine

Background:

• Increased prevalence of sleep disorders due to modern lifestyles impacts quality of life and work.
• Current polysomnography (PSG) methods for sleep disorder diagnosis are complex, bulky, and lack portability.
• There is a need for practical, non-invasive, and real-time sleep monitoring solutions.

Purpose of the Study:

• To propose an optical fiber sensor as a viable alternative for sleep monitoring.
• To introduce an optical fiber interferometer for capturing ballistocardiography (BCG) and electrocardiogram (ECG) signals.
• To develop and evaluate a novel machine learning method for sleep condition detection.

Main Methods:

• Development of an optical fiber sensor utilizing an interferometer.
• Integration of the sensor for non-invasive acquisition of BCG and ECG signals.
• Implementation of a new machine learning model for analyzing physiological signals and assessing sleep quality.

Main Results:

• The proposed optical fiber sensor demonstrates low power consumption and operates without interference.
• The system enables real-time health monitoring and accurate capture of BCG and ECG signals.
• Experimental results validate the superior performance of the sensor architecture and machine learning model in sleep quality assessment.

Conclusions:

• The optical fiber sensor presents a promising, practical solution for sleep monitoring, addressing limitations of conventional methods.
• The integrated machine learning approach enhances the accuracy of sleep condition detection and quality assessment.
• This technology facilitates improved management and understanding of sleep disorders.