Research
Brain
Actuated Technologies Inc. (BAT)
A company with a strong emphasis on Research
& Development
Founded by Dr. Andrew Junker, BAT began as an R&D company. Working for the United States Air Force for twenty years as a research scientist and also serving as a past director of the Human Engineering Program at Wright State University , set the stage for the skills BAT has in the R&D arena. Accordingly, BAT has been awarded Phase I and Phase II SBIR grants from the USAF and two Phase I SBIR grants from National Institutes of Health.
In addition to directing research of our own, BAT is dedicated to helping others engage in research with the Cyberlink. Towards this end we have created a Software Developer’s Kit (SDK) that allows researchers and developers to create their own software. We are also interested in entering into cooperative research efforts with others.
Please contact us if you would like further information about obtaining the SDK and/or if you are interested in undertaking joint research efforts with us. Below we present current R&D work being performed by BAT along with past project results and some of the work that has been done by other researchers working with the Cyberlink and SDK.
NIH Research and Development Effort by BAT:
Title: Coherent Detected Periodic Brainwave Computer
Control
This study was conducted under the National Institute of
Child Health and Human Development Grant Number R43 HD42942-01, of the National
Institutes of Health.
Abstract Of Research Plan:
The objective of this study was to demonstrate that users with severe muscle control disabilities could use forehead brainwave frequency bio-potentials to obtain computer access by controlling the Brainfingers system with their forehead derived bio-potentials. Four subjects with multiple disabilities participated in this study. They ranged in age from nine to twenty, attended one public special education school, and were restricted in their ability to access curriculum, communication, or leisure activities at the start of this study. At the beginning of the study none of the subjects had meaningful access to switch devices to facilitate communication, curriculum participation, or leisure activities. By the end of the study, the subjects were able to effectively control computers by using their Brainfingers derived brainwave frequency signals at their forehead. The results of this study indicate these subjects achieved effective switch and mouse access by controlling a computer through the Brainfingers system. The subjects reached a level of expertise with Brainfingers that allowed them to interact with the computer and operate software independently from the researcher. During the course of the study, functional and behavioral changes were observed from the four subjects. Likewise, all four learned the Brainfingers system at the same rate even though the experimental protocol was designed to allow independent learning rates.
A second objective was to employ coherent detection of periodic brain waves, taking advantage of the ability of a coherent detection system to reject interference from non-periodic muscle and random brainwave signals. The Brainfingers computer control system was used as the basic building block. Software was modified to operate with full coherent detection. A conventional 2nd order phase lock loop was selected as the design, where the signals were to be hard limited in order to achieve normalized operation. This design was predicated on the idea that periodic alpha wave signals could be generated by brain wave activity at the control of a disabled person and that interfering muscle signals had no periodic component and therefore would not be detected by a phase lock loop. It was discovered that even strong alpha waves were not generated as continuous waves, but rather as short bursts of about 1 second in duration, and separated by 2 or 3 seconds. Each burst was at a slightly different frequency, and there appeared to be no coherence between each burst in a series of bursts, thus negating the basic idea of a phase lock loop. Therefore, the phase lock loop approach to coherent processing was abandoned. Instead a lock-in amplifier approach was used to extract brainwave frequency signals from the forehead.
Results:
The performance results of the subjects exceeded the researchers' expectations. The four subjects identified with multiple disabilities achieved computer and software access through hands-free control of Brainfingers in less than sixteen hours over eight weeks. Results suggest that the Brainfingers system could satisfy the needs of persons identified with multiple disabilities who have physical limitations that interfere with computer operations. Brainfingers was found to be safe to use, easily learned, not affected by involuntary body, facial, and/or head movements, or limited vision, and offered specialized and universal computer and software access. For these subjects with multiple disabilities the brainwave frequency bands were better suited for control of up/down and clicking in the computer.
It is concluded that the Brainfingers system removed access barriers to computers and enhanced the subjects' quality of life by permitting them to operate software for communication, learning, and leisure, and enhanced their abilities to enjoy an independent lifestyle. Brainfingers allowed the subjects to advance their cognitive and social skills. Brainfingers provided the speed needed to use computer technology, made choices available, grew with the individual's development, and could be used in many environments.
Download Final Report:
Click on the following link to download a copy of the final report for this study > BATFinalNIH1.PDF.
NIH Research and Development Effort by BAT:
Title: Hands-free Computer Access for the
Severely Disabled: A study was conducted under the National Institute of Child
Health and Human Development; Grant Number R43 HD39070 of the National Institutes
of Health.
Abstract Of Research Effort:
The feasibility of using the Cyberlink hands-free computer controller as an access solution for individuals with disabilities was investigated. Specific questions evaluated were: (1) Can the Cyberlink be used as a computer access solution for individuals with no other documented means of computer access? (2) Can Cyberlink control be made more accurate and reliable through the use of machine learning and adaptive control techniques? Clinical trials were started with thirty-two participants of varying ages and disabilities. Each participant had been unable to access technology due to physical limitations. Twenty-five of the 32 participants were able to exhibit some form of conscious computer control with the Cyberlink and participated in the complete clinical trial design. Trials consisted of two tasks designed to measure response times and the ability to move a cursor to a target using the Cyberlink. Each task was completed by the participant a minimum of three times. History and performance data were collected and analyzed using Cyberlink software developed specifically for these tasks. Data was also obtained using questionnaires, videotapes and observation. Single trial test results indicate it is possible to recognize and discriminate participant actions using a window of time history to better recognize changes in response patterns. These findings indicate time series analysis of the data can be effective in distinguishing intended user response from background signal activity. In many cases, this permitted the recognition of the participant’s intention at an earlier time than a simple threshold crossing rule. This suggests implementing an adaptive control based upon time series analysis will be advantageous.
Summary of Results:
A total of 32 individuals with no other documented method of computer access were tested with the Cyberlink treatment to see if they could achieve computer access with the Cyberlink. Of these 32 individuals, 25 were found to be able to exhibit conscious control with the Cyberlink. These 25 individuals were selected as the participant population for this study. All 25 participants were able to complete at least one of the tasks of the Cyberlink treatment. Task completion can be interpreted as a demonstration of computer access, since all participants started with no documented form of computer access. Approximately 78% of the individuals tested, 25 out of 32, were able to achieve some form of access with the Cyberlink. These results are quite remarkable considering the profoundness of the disabilities of many of the participants.
Partners In Research:
This study was conducted by BAT with the help of the following Centers:
The Schiefelbusch Speech, Hearing and Language Clinic of the University of KansasProvidence Child Center, Portland Oregon
Success for Kids School, Loma Linda California
MDA ALS Center of Hope, MCP Hahnemann Univ., Philadelphia PA
Download Final Report:
Click
on the following link to download a copy of the final report for this study >
BATFinalNIH2.PDF .
Research and Development Efforts by Others:
Title: Cyberlink, Computer Access For Persons Identifed With Multiple Disabilities
by Danise Marie Marler Master of Arts in Special Education, CSUN
Data collected were from four subjects, identified with multiple disabilities, while they used a hands-free computer interface system called Cyberlink Brainfingers. They ranged in age from nine to twenty, attended one public special education school, and were restricted in their ability to access curriculum, communication, or leisure activities at the start of this study. The results of this study indicated that these subjects achieved effective switch and mouse access controlling the Cyberlink system. The subjects reached a level of expertise with the Cyberlink that allowed them to interact with a computer and operate software independently from the researcher. At the beginning of the study, the subjects were unable to create effective muscle control signals at their forehead, but by the end of the study, the subjects were able to effectively control the Cyberlink by using brainwave frequency signals at their forehead. During the course of the study, functional and behavioral changes were observed from the four subjects. Likewise, all four learned the Cyberlink Brainfingers system at the same rate even though the experimental protocol was designed to allow independent learning rates.
Click on the following link to download a copy of Dani Marler's Case Study: CyberlinkCaseStudy.pdf
| Robot Arm Control Work being done by Dr. Eamon Doherty
|
“Here
is a picture of Bruce using the Cyberlink with a robotic arm to lift and move
papers around the desk. The arm has a maximum lift of 6 ounces and moves at a
rate of about ½ inch per second. My website containing more information
about this work is www.headtrauma.com/sp2002.htm. Our next project is robotic
rovers that are controlled by the Cyberlink. We will then have Cyberlink radio
controlled devices that can be operated remotely by a person in a wheelchair for
example. One possible application would be for locating a care assistant or person
who wandered away from the institution.”
Mobile Robot Control By Brainwave Bio-Potentials
Choi
Kyoung ho, Gifu University, 1-1 Yanagido, Gifu
Minoru Sasaki, Gifu University
CHOI Kyoung ho and Minoru Sasaki are working on the development of a system whereby a person can control a mobile robot by using the components of his/her brain wave bio-potentials. Such a system may be used as a control device employing human eye-movements, facial muscle, and brain wave bio-potentials. As inputs, biofeedback is provided by indications on a computer screen so the human operator can practice his/her inputs before applying them to the control of a mobile robot. The operator's forehead bio-potentials can be acquired and processed in Cyberlink (TM) as mobile robot control source signals. The computer analyzes an operator's EEG (electroencephalographic), EOG (electrooculographic), and EMG (electromyographic) signals in real time. This analyzation is achieved by periodically checking EEG, EOG and EMG signals in real-time. The results are used to control a mobile robot. Operators performed successfully, with accuracy of the control improving with training. The experimental results suggest that a mobile robot can be operated by human brain wave bio-potentials.
Dr. Paul Gnanayutham,
Department of Computer and Information Sciences,
De Montfort University Milton Keynes,
Hammerwood Gate,
Kent Hill,
Milton Keynes MK7 6HP.
Email: pg@dmu.ac.uk
Dr. Gnanayutham’s main research efforts are researching intelligent communication interfaces for the disabled. The signals used for interfacing are EEG, EMG and EOG from the forehead. His research is designed to help people with disabilities communicate by special interfaces, using Artificial Intelligence to help them communicate easily. Dr. Gnanayutham is also working on a robotic interface to help persons with disabilities perform simple tasks. Research areas involved are Human Computer Interaction, Artificial Intelligence and Robotics. He presented an article “Robotics for the brain injured: An interface for the brain injured person to operate a robotic arm”, in October 2001, at the ICCIT'2001 conference, New York.” He demonstrated a simple robotic arm using the Cyberlink a and a simple Visual Basic program at the conference.
Please contact us for more information on R&D efforts.
Contact
Us: