A P300 Brain Computer Interface Keyboard to Control Assistive Technology for People With Amyotrophic Lateral Sclerosis
Overview
- Phase
- Not Applicable
- Intervention
- Not specified
- Conditions
- Amyotrophic Lateral Sclerosis
- Sponsor
- University of Michigan
- Enrollment
- 13
- Locations
- 1
- Primary Endpoint
- Accuracy of Typing With a BCI Keyboard by ALS Patients.
- Status
- Completed
- Last Updated
- 10 years ago
Overview
Brief Summary
People with Amyotrophic Lateral Sclerosis (ALS) will use a P300 based brain computer interface (BCI) keyboard to type in assistive technology devices. The results of this study will be compared with a previous study of a P300 BCI keyboard used by healthy volunteers.
Investigators
Jane Huggins, PhD
Research Assistant Professor
University of Michigan
Eligibility Criteria
Inclusion Criteria
- •Age 18 or older.
- •Diagnosed with cerebral palsy, spinal cord injury, neuromuscular disease, or ALS that results in impaired hand and arm function making it difficult to manipulate objects (if at all) and requires help to prepare or modify activities
- •Able to see the BCI display
- •Able to give informed consent
- •Able to understand and remember instructions concerning participation
- •Able to communicate effectively at least with familiar conversation partners
Exclusion Criteria
- •Are unable to give informed consent.
- •Are currently experiencing open head sores
- •Have a history of photo-sensitive epilepsy
- •Have a history of cognitive deficits requiring accommodation or special education services
- •Are unable to sit without moving the head and neck for at least 15 minutes
Outcomes
Primary Outcomes
Accuracy of Typing With a BCI Keyboard by ALS Patients.
Time Frame: 3 times over 2-4 weeks
Accuracy for the sentence typed in each environment was calculated as the percentage of characters for which the result character matched the target character. The target characters were determined based on the next character needed to complete the sentence to be copied. In the case of errors, the next character was therefore a backspace to correct the error. The target characters were modified by subject comments to account for errors in selecting the next character. Once sentence was typed in each environment in each session on a separate day. From the three repeated sessions, there were therefore 9 total sentences per subject with 3 measures for each environment. These were treated as repeated measures for the analysis.