Related Experiment Video
Updated: Jul 13, 2025
06:09
P300-Based Brain-Computer Interface Speller Performance Estimation with Classifier-Based Latency Estimation
Published on: September 8, 2023
611
A click-based electrocorticographic brain-computer interface enables long-term high-performance switch-scan spelling.
Nathan Crone1, Daniel Candrea2, Samyak Shah1
1Johns Hopkins Hospital.
Research Square
|October 16, 2023
Summary
Brain-computer interfaces (BCIs) enable neural control for communication. A single "click" decoder, trained with minimal data, provided sustained text-based communication for an individual with ALS over 90 days.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Rehabilitation Technology
Background:
- Brain-computer interfaces (BCIs) offer communication restoration for individuals with severe motor and speech impairments.
- Single-command "click" decoders represent a fundamental yet effective BCI capability.
Approach:
- A high-density electrocorticographic (ECoG) BCI system was implanted to cover the sensorimotor cortex in a participant with amyotrophic lateral sclerosis (ALS).
- A click decoder was trained using limited ECoG data (<44 minutes over 4 days) collected up to 21 days before use.
- The decoder's performance and stability were evaluated over 90 days without retraining.
Key Points:
- The participant achieved a median spelling rate of 10.2 characters per minute using the click decoder with a switch-scanning interface.
- Despite a temporary signal interruption, a newly trained decoder, using even less data (<15 minutes), demonstrated comparable performance.
- This highlights the robustness of ECoG-based click decoding.
Conclusions:
- A click decoder can be trained effectively with a small ECoG dataset.
- Robust performance is maintained over extended periods, enabling functional text-based communication for BCI users.
- This technology holds significant promise for individuals with communication disabilities.
