Non-Contact Detection of Apnea-like Breathing Cessations Using Laser Speckle Pattern Analysis

Area of Science:

• Biomedical Engineering
• Medical Diagnostics
• Sleep Medicine

Background:

• Sleep apnea is a common disorder impacting sleep quality.
• Current diagnostic methods like polysomnography (PSG) are costly and inconvenient.
• There is a need for accessible, non-contact sleep apnea screening solutions.

Purpose of the Study:

• To develop and evaluate a novel non-contact biosensing system for detecting apnea-like breathing cessations.
• To assess the system's accuracy using secondary laser speckle pattern analysis.
• To explore its potential as a low-cost, home-based screening tool.

Main Methods:

• A non-contact biosensing system employing secondary laser speckle pattern analysis was developed.
• Image processing algorithms were used to analyze breathing cessations.
• The system was tested on 14 healthy subjects simulating apnea-like events (10-30s durations) in various positions.
• Five-fold cross-validation (participant-level and class-level) was applied.

Main Results:

• The system achieved an overall accuracy of 87.0% ± 3.0% using class-wise stratification.
• Sensitivity for long cessations was 91% ± 12.4%, and for short cessations was 88% ± 11%.
• The two-class classification strategy confirmed the approach's robustness.

Conclusions:

• Secondary laser speckle pattern analysis shows promise for sleep apnea detection.
• The developed system offers a potential low-cost, non-contact alternative for home screening.
• This technology could improve accessibility to sleep apnea diagnosis.