Multimodal spatial calibration for accurately registering EEG sensor positions

Area of Science:

• Biomedical Engineering
• Neuroscience
• Computer Vision

Background:

• Accurate sensor positioning is crucial for electroencephalography (EEG) source localization.
• Existing methods for calibrating multiple multimodal sensors can be time-consuming and complex.

Purpose of the Study:

• To develop a fast and accurate photogrammetry-based calibration method for multiple multimodal sensors, specifically for EEG sensor localization.
• To enable simultaneous acquisition of all EEG sensor positions on the human head.

Main Methods:

• A novel photogrammetry system was developed for rapid localization of EEG sensors.
• A multiple views' calibration process was implemented to determine sensor transformations.
• An efficient local repair algorithm was used to enhance depth map accuracy.
• A specialized calibration body was designed to facilitate the calibration process.

Main Results:

• The proposed method achieved accurate and robust calibration results.
• Evaluation using a chessboard calibration plate demonstrated good performance.
• The system successfully obtained simultaneous EEG sensor positions.

Conclusions:

• The developed photogrammetry system offers a fast and accurate solution for multi-sensor calibration.
• This method is well-suited for applications requiring precise EEG sensor localization, such as in human brain source localization.
• The approach enhances the feasibility of complex neuroimaging studies.