Intelligent Method for Real-Time Portable EEG Artifact Annotation in Semiconstrained Environment Based on Computer
Xuesheng Qian1,2, Mianjie Wang3, Xinyue Wang4
1Institute of Systems Engineering and Collaborative Laboratory for Intelligent Science and Systems, Macau University of Science and Technology, Macao 999078, China.
Computational Intelligence and Neuroscience
|February 22, 2022
Summary
Portable EEG technology (PEEGT) faces challenges with artifacts in natural settings. This study introduces a computer vision method to identify and annotate EEG artifacts from participant behavior in real-time, enabling more accurate neurotesting.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Computer Vision
Background:
- Portable EEG technology (PEEGT) offers potential for real-world neuroscience research.
- Artifacts from subject activities in semi-constrained environments limit PEEGT's application.
- Current artifact annotation methods hinder PEEGT's portability and cost-effectiveness.
Purpose of the Study:
- To develop an intelligent method for real-time identification and annotation of EEG artifacts.
- To address limitations in current postprocessing artifact removal techniques.
- To enable large-scale neurotesting in natural environments outside the lab.
Main Methods:
- Utilized computer vision (CV) to detect participant blinks and head movements, key sources of EEG artifacts.
- Developed a real-time artifact annotation system based on recognized participant behaviors.
- Shifted from signal-based postprocessing to behavior-based preprocessing for artifact management.
Main Results:
- The CV-based method effectively identifies and annotates EEG artifact segments in real-time.
- Comparative experiments validated the effectiveness of the CV method against manual annotation.
- The approach lays the groundwork for accurate, real-time artifact removal in PEEGT.
Conclusions:
- Computer vision offers a viable solution for real-time EEG artifact management in natural environments.
- This method facilitates cost-effective, large-scale neurotesting without expensive lab equipment.
- Enables wider adoption of PEEGT in real-world neuroscience applications.
