2002 Conference Proceedings

Go to previous article 
Go to next article 
Return to 2002 Table of Contents


CYBERLINK BRAINFINGERS FOR PEOPLE WITH NO MEANS OF ACCESS, NIH STUDY RESULTS

Dr. Andrew M. Junker, Brain Actuated Technologies Inc.
Email: admin@brainfingers.com 

Dr. Jane R. Wegner, University of Kansas
Email: Janew@dole.isi.ukans.edu 

Dr Thomas Sudkamp, Wright State University
Email: tsudkamp@cs.wright.edu

SUMMARY

Results of a study to test the potential of using Cyberlink Brainfingers as a hands-free alternative and augmentative communication (AAC) access device for twenty-five individuals with no other means of access will be presented.

INTRODUCTION

Results of a study to test the potential of using Cyberlink Brainfingers as a hands-free alternative and augmentative communication (AAC) access device for twenty-five individuals with no other means of access will be presented. This study was conducted under the National Institutes of Health, National Institute of Child Health and Human Development Grant Number 1 R43 HD39070-1. The first year of the study was devoted to developing needed Cyberlink system capabilities. Results of this development effort were presented at CSUN 2001. Making use of developed system, an intervention study was conducted with twenty-five individuals of varying ages and disabilities who had been unable to access AAC technology due to physical limitations. Subjects were presented with various Cyberlink tasks that were designed to evaluate the potential for AAC. Results of this intervention study will be presented.

SIGNIFICANCE

There are individuals, who because of the severity of their physical limitations have been unable to access AAC technology through either direct selection or scanning via a switch. These individuals often have disabilities related to cerebral palsy (CP), amyotrophic lateral sclerosis (ALS), multiple sclerosis (MS), muscular dystrophy (MD), or traumatic brain injury (TBI).

Other practical barriers involve physical fatigue associated with use, length and intensity of required system calibration and adjustment, length and intensity of required user training. Because of the access difficulties, communication for these individuals is often limited, left to interpretations by communication partners, or non-existent. Given the growth and advances in AAC technology that are taking place, a viable alternative access is the only factor standing between individuals with the most severe disabilities and the technology available to provide independent communication. Cyberlink Brainfingers has the potential to bridge the gap between these individuals and the communication technology available to empower and transform their lives.

CYBERLINK BRAINFINGERS

The Cyberlink Brainfingers Controller is a device that responds to forehead voltages (bio-potentials) resulting from subtle facial muscle, brainwave and eye movement activity. The Cyberlink user is able to point and click a mouse, access the Internet, and control third party software such as EZ-Keys, Clicker 4, Gus and Wivik, totally hands-free. As a recreational device, it is also possible to use the Cyberlink to play interactive video games and music. Access for people with disabilities looks very promising. The Cyberlink potential is perhaps best summed up by an email from a current Cyberlink user. In this email he wrote; "I am so busy writing emails that I forget I am sick." He used the Cyberlink with EZ-Keys to write this email. He has ALS.

The Cyberlink user wears a thin, lightweight, cloth headband containing three plastic sensors that detect the voltages found at the forehead. These sensors are electrically conductive plastic, reusable, inexpensive to replace, and have been found to last for more than three months of continuous use. The headband connects to a Cyberlink Interface box that amplifies and digitizes the forehead voltage into three signals. These three signals are an eye movement signal, a brain-body signal and a muscle signal. The brain-body signal (in the range 1 Hz to 45 Hz) contains both brainwave and body muscle bio-potentials. The three signals are sent to a PC computer through a serial port interface.

The Cyberlink software, installed in the PC, further amplifies and decodes these signals. The combined amplification of the Cyberlink Interface box and PC software is over one million. Because of this sensitivity the Cyberlink is responsive to the subtlest of muscle activity as well as brainwaves.

The decoding process creates eleven continuous and 4 discrete signals. Any two of the eleven continuous signals can be used to control the mouse cursor. The lower ten continuous signals span the accepted EEG theta, alpha and beta frequencies. The 11th continuous signal is derived from the "muscle only" signal. The 4 discrete signals (an eyebrow lift, a jaw close, a glance to the left, a glance to the right) can be used to control mouse button clicks and keystrokes. Since the Cyberlink moves the point of control from the user's fingers to the forehead and thus closer to the user's brain, we call the control signals created by the Cyberlink, "Brainfingers".

The Cyberlink software teaches the user how to bring their forehead voltages under conscious control. The software allows for easy adjustment and reconfiguration of the Cyberlink to accommodate to the current state of the user's control capabilities. The software allows the user to choose what Cyberlink signals are to be mapped to left/right and up/down cursor movement, left mouse click, right mouse click and specific keyboard key presses. Based upon this mapping, user generated Cyberlink signals trigger the Cyberlink software to send Windows messages to the appropriate Windows computer software.

When a person uses the Cyberlink for the first time a new user file is created. To start, default settings are provided. As the user works with the Cyberlink and makes changes to his or her settings the changes are saved within the user's file.

The startup default settings provide the most intuitive and easily controlled Cyberlink configuration. This includes mapping a short facial muscle signal to a mouse click or key press, mapping a gradual increasing/decreasing of the facial muscle signal to up/down cursor motion, and mapping a left/right glance to left/right cursor motion. In this mode of control users are able to achieve control of the cursor in a very short time. Users have found that they could use the initial startup configuration when their disability prevented them from using any other access device.

Achieving click control of scanning third party software such as an on-screen keyboard like EZ-Keys, is possible within minutes. Most third party on-screen keyboards in a scan/click mode can be controlled by a mouse click or some keyboard key press. For example, a right mouse click controls EZ-Keys in scan/click mode, a left arrow key press can control Clicker 4, an F12 key press can control Wivik, and a left mouse click can control Gus. Because the Cyberlink software allows the user to map a click gesture to a left or right mouse click or keyboard key press, the user is able to control these on-screen keyboards. It is also possible to use the Cyberlink software to toggle between full mouse mode for surfing the net etc. and scan/clicking for text entry.

If a person's disability prevents them from using the initial default configuration it is easy to reconfigure the Cyberlink software. Often it is a matter of just increasing the sensitivity of the user's signal, or adjusting for a delayed response, or adjusting for a persistent response. Due to the graphical interface it is easy to make the needed adjustments. Since the Cyberlink software can be controlled completely hands-free the user can make his or her adjustments without assistance.

The startup default configuration is the easiest and quickest to learn and is the preferred mode of control. Other modes of control are possible with the Cyberlink. Depending upon the severity of the disability, click control as opposed to full mouse control may be the best option for the user. Clicking with the Cyberlink is possible with any one of the 11 Brainfingers. If the muscle Brainfinger won't work it is possible to use one of the other 10 Brainfingers.

The three lowest frequency Brainfingers are sensitive to left/right eye movements. When no other means of control is possible, Cyberlink users are able to use one of these low frequency Brainfingers for click control of an on-screen scanning keyboard.

The middle three Brainfingers are sensitive to alpha brainwaves. Learning this kind of control takes longer to master. Initially alpha Brainfinger control requires relaxing of one's thoughts and softening of one's visual focus to make the alpha Brainfinger signals increase. Learning alpha Brainfinger control is like learning a new motor skill and thus takes practice and time to learn.

Four of the Brainfinger signals are in the beta brainwave region. An increase in beta energy is usually associated with increased mental focus/activity. Brainfingers in the beta frequency range are also sensitive to low frequency muscle activity. Thus a good way to learn beta Brainfinger control is by performing some sort of intense mental activity coupled with the subtlest of muscle activity such as imagining/trying to contract the top of the scalp. The beta Brainfingers can be adjusted to the resulting small signal changes. Learning beta Brainfinger control takes time and practice.

Once consistent Cyberlink control is mastered, the user can control the mouse cursor and/or mouse click to access an on-screen keyboard such as EZ-Keys, Reach, Gus, Wivik, KeyWi or Clicker 4. Thus the Cyberlink Brainfingers user will be able to surf the net, do email and anything else that can be done on a PC with a hand-mouse.

RESULTS TO BE PRESENTED

Results of this study will be presented. The potential benefits from the incorporation of the learning and adaptive control algorithms will be discussed. Focus will be in terms of the reduction of the length and intensity of user training, the development of a system more tailored to the capabilities of the individual, and the development of a system that will adapt to compensate for changes in the user's performance.


Go to previous article 
Go to next article 
Return to 2002 Table of Contents 
Return to Table of Proceedings


Reprinted with author(s) permission. Author(s) retain copyright.