A visual brain-computer interface as communication aid for patients with amyotrophic lateral sclerosis

Area of Science:

• Neuroscience
• Biomedical Engineering
• Rehabilitation Technology

Background:

• Amyotrophic Lateral Sclerosis (ALS) progressively impairs motor function, necessitating assistive communication technologies.
• Existing visual Brain-Computer Interface (BCI) spellers have limitations in speed and accuracy for ALS patients.
• Code-modulated Visual Evoked Potentials (cVEP) offer a potential advancement for visual BCI spellers.

Purpose of the Study:

• To evaluate the brain responses, Brain-Computer Interface (BCI) performance, and user experience of a cVEP speller in healthy individuals and ALS patients.
• To compare the efficacy of the cVEP speller against existing visual BCI technologies for ALS communication.

Main Methods:

• A cVEP speller was tested on 20 healthy participants and 10 individuals with ALS.
• Participants engaged in cued and free spelling tasks, as well as Yes/No answer selections.
• Brain responses, BCI performance metrics, and user experience were systematically assessed.

Main Results:

• The cVEP speller achieved high accuracy across all participant groups: 79% for ALS patients, 88% for older healthy adults, and 94% for younger healthy adults in cued spelling.
• All participants successfully answered Yes/No questions with approximately 90% accuracy.
• ALS patients achieved an average typing speed of 10 characters per minute.

Conclusions:

• The cVEP speller demonstrates superior performance compared to other visual BCI spellers for ALS patients.
• The study confirms the general usability of cVEP signals for ALS patients, indicating potential applications beyond spelling, such as smart home device control.