Related Experiment Video
Updated: Aug 2, 2025
07:54
Drug-Induced Sleep Endoscopy DISE with Target Controlled Infusion TCI and Bispectral Analysis in Obstructive Sleep Apnea
Published on: December 6, 2016
19.9K
Automatic identifying OSAHS patients and simple snorers based on Gaussian mixture models
Xiaoran Sun1, Li Ding1, Yujun Song1
1School of Physics and Optoelectronics, South China University of Technology, Guangzhou, 510640, People's Republic of China.
Physiological Measurement
|April 14, 2023
Summary
This study developed an effective system to detect Obstructive Sleep Apnea Hypopnea Syndrome (OSAHS) using snoring sounds. The Gaussian mixture model achieved 90% accuracy, offering a low-cost, home-based diagnostic tool.
Area of Science:
- Biomedical Engineering
- Sleep Medicine
- Acoustic Signal Processing
Background:
- Snoring is a common symptom of Obstructive Sleep Apnea Hypopnea Syndrome (OSAHS).
- Accurate and accessible detection methods for OSAHS are crucial for timely intervention.
- Distinguishing simple snoring from OSAHS-related snoring is essential for diagnosis.
Purpose of the Study:
- To develop an effective system for detecting OSAHS patients based on snoring sounds.
- To utilize acoustic characteristics of snoring for differentiating simple snorers from OSAHS patients.
- To evaluate the performance and efficiency of the proposed detection model.
Main Methods:
- Employing Gaussian Mixture Models (GMM) to analyze nocturnal snoring sound characteristics.
- Selecting acoustic features based on the Fisher ratio for model training.
- Conducting a leave-one-subject-out cross-validation experiment with 30 participants (6 simple snorers, 24 OSAHS patients).
Main Results:
- Snoring sounds from simple snorers and OSAHS patients exhibit distinct distribution characteristics.
- The GMM-based model achieved an average accuracy of 90.0% and precision of 95.7% with 100 selected features.
- The average prediction time for the model was 0.134 ± 0.005 seconds.
Conclusions:
- The proposed system effectively diagnoses OSAHS patients using snoring sounds.
- The method demonstrates a low computational cost, making it suitable for home-based diagnosis.
- Snoring sound analysis presents a promising, non-invasive approach for OSAHS detection.
