Related Experiment Video
Updated: Jan 8, 2026
06:14
The Combination of Transcranial Alternating Current Stimulation and Electroencephalogram
Published on: October 10, 2025
435
Ocular artifact from electroencephalogram - a comparative analysis of feature extraction, selection and
Malika Garg1, Jasbir Kaur1, Neelam Rup Prakash1
1Department of Electronics and Communication, PEC (Deemed to be University), Chandigarh, India.
Journal of Medical Engineering & Technology
|December 17, 2025
Summary
This study benchmarks methods for removing eye blink artefacts from electroencephalogram (EEG) signals. The best systems achieved 93.8% accuracy, improving EEG analysis for clinical and brain-computer interface applications.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Signal Processing
Background:
- Electroencephalogram (EEG) signals record brain activity but are often contaminated by artefacts.
- Ocular artefacts, particularly eye blinks, significantly hinder EEG signal analysis.
- Accurate artefact removal is crucial for reliable EEG interpretation in clinical and research settings.
Purpose of the Study:
- To conduct a comparative analysis of various techniques for classifying and removing ocular artefacts from EEG signals.
- To evaluate the performance of different feature extraction, feature selection, and classification methods for artefact identification.
- To establish a benchmark for ocular artefact identification accuracy in EEG data.
Main Methods:
- Utilized three distinct feature extraction methods on EEG recordings from eight subjects.
- Applied three feature selection algorithms and 30 classification methods.
- Evaluated system performance using 5-fold cross-validation, calculating accuracy across 360 feature-classifier combinations.
Main Results:
- The highest accuracy achieved was 93.8%.
- Optimal performance was obtained using wavelet-based features and principal component analysis for feature selection.
- Kernel Naïve Bayes, Linear Support Vector Machine (SVM), and Ensemble Bagged Trees were the top-performing classifiers.
Conclusions:
- This comprehensive benchmark demonstrates exceptional accuracy in ocular artefact identification.
- The validated methods show significant potential for real-time EEG preprocessing.
- Improved EEG signal quality can enhance clinical diagnostics and Brain-Computer Interface (BCI) applications.
