Files and links (1)
PDFOpen Access (License Unspecified), Open Access
Thesis
Feasibility and optimization of a P300-based brain computer interface in individuals with amyotrophic lateral sclerosis
Master of Science (M.S.), Drexel University
Jun 2009
DOI:
https://doi.org/10.17918/etd-3062
Abstract
Amyotrophic Lateral Sclerosis is a neuromuscular disease characterized by progressive weakness resulting in a state of profound disability including the loss of functional speech. The rise of new technologies allows people living with ALS and other individuals with severe motor disabilities to communicate using alternate methods. One alternative communication method is an Electroencephalographic (EEG) based brain-computer interface (BCI), which uses a cap embedded with electrodes to read EEG signals. In particular, the P300, a naïve response to stimuli, is used. Through a P300 Speller paradigm, the EEG-based BCI allows individuals with severe disabilities to communicate using a computer even when conventional devices that require mechanical manipulation have failed. An electrode test system was designed to determine whether the commercial electrode cap was functioning correctly. Direct input from a function generator was provided to 4 electrodes at a time and the resulting signal was measured using a DATAQ acquisition box and signal acquisition software. A generated sine was seen in each electrode with a signal loss of 5-6%. The electrodes were able to adjust to and reflect changes in input amplitude and frequency, demonstrating adequacy in signal acquisition. Four able-bodied and eight individuals with ALS from the Philadelphia community participated in the P300 Speller trials under informed consent to determine the feasibility of using the BCI in an ALS population. The EEG was recorded with 8 electrodes using an electrode cap. All aspects of data collection were controlled by the BCI2000 system. Users were asked to participate in a copy-spelling session in which they attended to a specified target letter appearing in a letter matrix. All controls and 6 out of 8 individuals with ALS were considered to be responders (spelling accuracy over 75%). Spelling ability is not correlated to the ALS Functional Rating Scale (ALSFRS), age, or gender. This indicates that individuals who are extremely disabled are able to accurately use a BCI. There are differences in the P300 signal between healthy controls and individuals with ALS. The latency of the peak amplitude of the P300 signal is significantly (p=0.020) later in healthy controls compared to individuals with ALS. The peak amplitude of the P300 signal is not significantly different in healthy controls compared to individuals with ALS. In 3 out of 4 healthy controls, activity can be visualized across all 8 electrodes in the cap whereas in 7 out of 8 individuals with ALS, activity can be visualized primarily in channels 1-4. Changes in latency and signal movement through the electrodes may indicate differences in the electrical wiring in the brain. However, these changes do not affect the ability of an individual to use the BCI and do not influence the amplitude of the signal. The ground and reference electrode locations were changed to determine the flexibility of the BCI and to optimize electrode placement. Examined healthy controls and individuals with ALS were considered responders at each electrode location. The ideal location and number of flashing sequences varies between individuals, however, the ability to move the electrodes without detriment demonstrates that the system can be manipulated to improve comfort and overall satisfaction with the BCI. BCIs can be used by individuals with a debilitating disease such as ALS to communicate with the external world and control their environment. The BCI system and the P300 Speller paradigm are dynamic, flexible, and can be made to work for the majority of individuals with both comfort and ease.
Metrics
39 File views/ downloads
22 Record Views
Details
- Title
- Feasibility and optimization of a P300-based brain computer interface in individuals with amyotrophic lateral sclerosis
- Creators
- Jennifer Giles King - DU
- Contributors
- Ryszard Lec (Advisor) - Drexel University (1970-)
- Awarding Institution
- Drexel University
- Degree Awarded
- Master of Science (M.S.)
- Publisher
- Drexel University; Philadelphia, Pennsylvania
- Resource Type
- Thesis
- Language
- English
- Academic Unit
- School of Biomedical Engineering, Science, and Health Systems (1997-2026); Drexel University
- Other Identifier
- 3062; 991014632840304721