Related Experiment Video
Updated: Jul 2, 2025
09:42
Acquisition and Semi-Automated Analysis of Respiratory Muscle Surface Electromyography
Published on: January 24, 2025
530
Computational ensemble expert system classification for the recognition of bruxism using physiological signals
Pragati Tripathi1, M A Ansari1, Tapan Kumar Gandhi2
1Department of Electrical Engineering, Gautam Buddha University, Greater Noida, India.
Heliyon
|February 23, 2024
Summary
This study developed an automatic diagnostic system for bruxism (teeth grinding) using machine learning. Combining EEG, ECG, and EMG signals achieved 99% accuracy, improving diagnosis for this sleep disorder.
Area of Science:
- Biomedical Engineering
- Sleep Medicine
- Machine Learning
Background:
- Bruxism, a sleep-related disorder involving teeth grinding and clenching, poses diagnostic challenges with current methods.
- Existing diagnostic approaches for bruxism are often inefficient and difficult to implement.
- There is a need for improved, automated diagnostic schemes for effective bruxism management.
Purpose of the Study:
- To develop an advanced automatic diagnostic scheme for bruxism.
- To enhance the accuracy and efficiency of bruxism diagnosis using biological signals.
- To investigate the efficacy of a novel hybrid machine learning classifier for sleep disorder diagnosis.
Main Methods:
- Utilized a hybrid machine learning classifier within the Weka tool for bruxism diagnosis.
- Processed biological signals, including Electroencephalography (EEG), Electrocardiography (ECG), and Electromyography (EMG), by calculating power spectral density.
- Classified sleep stages (wake and REM) and bruxism presence using specific physiological signal channels.
Main Results:
- EEG-based diagnosis achieved 93% specificity and 95% accuracy.
- ECG-based classification yielded 87% specificity and 96% accuracy.
- Combining EEG, ECG, and EMG signals with an ensemble Weka tool resulted in the highest diagnostic performance: 97% specificity and 99% accuracy.
Conclusions:
- Integrating multiple physiological signals (EEG, ECG, EMG) significantly enhances bruxism diagnosis precision.
- The proposed machine learning approach offers a substantial improvement in diagnostic accuracy for bruxism.
- The developed method holds potential for improving automatic home monitoring systems for sleep-related disorders like bruxism.
