Topological Data Analysis of Electroencephalogram Signals for Pediatric Obstructive Sleep Apnea
Summary
Topological data analysis reveals distinct brain network differences in children with obstructive sleep apnea (OSA). This method shows potential for easier OSA diagnosis using electroencephalogram (EEG) data.
Area of Science:
- Neuroscience
- Computational Biology
- Signal Processing
Background:
- Obstructive sleep apnea (OSA) is a prevalent condition in children.
- Current diagnostic methods like polysomnography are resource-intensive.
- Understanding brain network alterations in pediatric OSA is crucial.
Purpose of the Study:
- To investigate the utility of Topological Data Analysis (TDA) for differentiating brain connectivity networks between pediatric OSA patients and controls.
- To explore TDA's potential as a scalable diagnostic tool for OSA.
Main Methods:
- Brain connectivity networks were derived from electroencephalogram (EEG) signals.
- Topological Data Analysis (TDA) was applied to these networks.
- Statistical comparisons were made between groups of pediatric patients with and without OSA.
Main Results:
- TDA identified statistically significant differences in brain dynamics between pediatric OSA patients and controls.
- The analysis demonstrated the robustness of TDA in detecting these differences even with noisy EEG data.
Conclusions:
- Topological Data Analysis shows promise as a non-invasive tool for identifying obstructive sleep apnea in children.
- This approach could lead to more accessible and scalable diagnostic methods for pediatric OSA, potentially reducing reliance on full polysomnography.
