Electrogastrogram-based detection of cybersickness with the application of wavelet transformation and machine learning: A case study
Introduction/purpose: The application of virtual reality (VR) and simulation technologies in military training offers cost-effective and versatile approach to training enhancement. However, prevalence of cybersickness (CS), characterized by symptoms such as nausea, limits their widespread use. Metho...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
University of Defence in Belgrade
2025-01-01
|
| Series: | Vojnotehnički Glasnik |
| Subjects: | |
| Online Access: | https://scindeks.ceon.rs/article.aspx?artid=0042-84692501079T |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Introduction/purpose: The application of virtual reality (VR) and simulation technologies in military training offers cost-effective and versatile approach to training enhancement. However, prevalence of cybersickness (CS), characterized by symptoms such as nausea, limits their widespread use. Methods: This study introduces objective parameters for the detection of CS using three-channel electrogastrogram (EGG) recording from one specific subject and assesses the independence and linear correlation for appropriate channel selection. The paper employs a 3-level discrete wavelet transformation (DWT) on the chosen channel to identify key parameters indicative of gastric disturbances. Furthermore, the paper investigates recovery from CS following VR and examines the application of unsupervised machine learning (ML) for segmenting EGG into baseline and CS, utilizing significant features previously identified. Results and discussion: The analysis reveals no significant differences across EGG channels and moderate to low linear correlation between channel pairs. The feature selection demonstrates that the root mean square of the amplitude as well as the maximum and mean values of the power spectral density (PSD) calculated on all DWT coefficients, are effective for CS detection while the dominant EGG scale could not indicate CS for any level of decomposition. Furthermore, recovery signs appear approximately 8 minutes after the first VR experience supporting the idea of conducting multiple sessions the same day i.e., intensive VR-based training. Conclusions: The unsupervised ML shows potential in identifying CSaffected EGG signal segments with feature extraction based on DWT, offering a novel approach for enhancing the prevention of CS occurrence in VR-based military training and other VR-related environments. |
|---|---|
| ISSN: | 0042-8469 2217-4753 |