Related Experiment Video
Updated: Aug 8, 2026
07:52
Multiscale Investigations of Cortical Processing by Integrating Laminar Polytrodes and Optogenetics with Micro Electrocorticography in Rodents
Published on: May 23, 2025
Optimal Size of Electrocorticography Grids for Classification of Hand Movements
Elena C Offenberg1, Dirk Keller1,2, Siamak Mehrkanoon2
1Department of Neurology and Neurosurgery, University Medical Center Utrecht Brain Center, Utrecht University, Utrecht, The Netherlands.
Neuroinformatics
|August 7, 2026
Summary
Researchers found that smaller electrocorticography (ECoG) grids can effectively decode hand movements for brain-computer interfaces (BCIs). Optimizing electrode placement significantly reduces grid size without impacting BCI performance, enabling safer implants.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Rehabilitation Technology
Background:
- Implanted Brain-Computer Interfaces (BCIs) offer potential alternatives to traditional assistive devices for individuals with severe motor impairments.
- Advancements in BCI performance are nearing clinical viability, but larger electrocorticography (ECoG) electrode grids increase surgical burden, hindering adoption.
Purpose of the Study:
- To investigate the impact of reducing ECoG grid size on hand movement classification performance.
- To determine the minimum ECoG grid area required for effective BCI function.
Main Methods:
- Explored hand movement classification (4, 5, 8 classes) using various rectangular ECoG subgrids (32, 64, 128 channels) in nine epilepsy patients.
- Analyzed classification performance across different subgrid sizes to identify optimal electrode placement strategies.
Main Results:
- ECoG grid surface area could be reduced by 75-94% with minimal impact on classification performance by focusing on informative areas.
- Classification performance remained stable until subgrids reached approximately 60mm², after which it declined significantly.
- Smallest subgrids above the threshold achieved comparable F1 scores (81.64-99.71%) to full grids (>230mm², 82.85-96.75%).
Conclusions:
- ECoG-based BCIs can accurately decode multiple hand movements using smaller, strategically placed electrode grids.
- This research supports the development of smaller, safer BCI implants, potentially accelerating clinical adoption for assistive technologies.
