BCI Devices Open Doors for People with Disabilities
February 2009 Issue
The power of the brain is being harnessed in astonishing ways, and the implications for people with limb loss, paralysis, an inability to speak, and other disabilities is exciting, to say the least. Brain-computer-interface (BCI) technology is opening doors to using the power of the mind to overcome the limitations of the body.
Imagine being able to make something move halfway around the globe simply by thinking about it. This feat was recently accomplished by a monkey-with the help of researchers at Duke University Medical Center, Durham, North Carolina.
|BCI2000 consists of a signal acquisition module that acquires brain signals from g.USBamp or g.MOBIlab+ devices. These raw signals are visualized and stored to disk and then submitted to the signal-processing module. The signal-processing module extracts signal features and translates them into device commands. These commands are used by the applications module to generate action. Image courtesy of Dr. Jonathan Wolpaw, Wadsworth Center, Albany, NY.|
The monkey's thoughts were able to control the walking patterns of a robot in Japan. To explore brain activity and record data, sensors tracked walking patterns while two rhesus monkeys walked forward and backward at different paces on a treadmill and researchers analyzed the relationship between leg movement and activity in the brain's motor and sensory cortex. The researchers found that certain neurons in different areas of the brain fire at different phases and at varying frequencies, depending on their role in controlling the complex, multi-muscle process of motion, according to senior research investigator Miguel Nicolelis, MD, PhD, quoted in an article by K.C. Jones in Information Week, January 16, 2008.
Researchers recorded brain activity, predicted the pattern of locomotion, and sent the signal from the motor commands of one of the animals to the robot in real time, he said.
"They can walk in complete synchronization," Nicolelis said. "The most stunning finding is that when we stopped the treadmill and the monkey ceased to move its legs, it was able to sustain the locomotion of the robot for a few minutes-just by thinking- using only the visual feedback of the robot in Japan."
The research expanded on previous experiments in Nicolelis' laboratory that showed monkeys could control the reaching and grasping movements of a robotic arm with their brain signals.
Surprising Portal to the Brain
Whether to control the movement of a cursor on a computer screen or the complex movements of prosthetic limbs, the brain not only needs to send signals, but it also needs to receive input from the environment to determine if it is directing the moves correctly to achieve the desired results. Currently, most of the feedback is visual. "We would like to get more sensory feedback as well, and we are looking at minimally invasive ways to do this," says Justin Williams, PhD, assistant professor, Department of Biomedical Engineering, Department of Neurological Surgery, University of Wisconsin-Madison.
A surprising organ-the tongue-is proving to be a portal to the brain. "People think this is odd when I first tell them," says Williams, "but the tongue is an excellent place to stimulate. It is very wet, so there is no problem with conductivity, and it is highly innervated with lots of different types of nerves."
He adds, "People can use brain signals to move a cursor on a screen blindfolded; they just get the location of the cursor through their tongue." A modality which could perhaps be used to provide information to the brain about heaviness, texture, slipperiness, and other characteristics as an adjunct to visual feedback would be very helpful in using prosthetic arms and hands to grasp, hold, and manipulate objects-information we generally receive through our fingertips.
Williams referenced an article in the MIT Technology Review , November 24, 2008, that discussed this research. The article, "Tongue Control: Sensory Feedback via the Tongue Might Improve Neural Prostheses," by Emily Singer, described the tongue stimulator as a thin-film array of 144 electrodes, slightly larger than a quarter, that sits on the tongue. "A stimulator delivers electrical signals based on visual information-in this case, the movement of a dot on a computer screen," Singer explains. "A similar device is already in use for people with balance disorders-tongue stimulation tells the user whether her head is upright-and is also being tested as a visual aid for the blind."
Defining a BCI
A brain-computer interface (BCI), also known as a direct neural interface or a brain-machine interface, creates a direct communication pathway between a brain and an external device. The differences between BCIs and neuroprosthetics are mostly in the ways the terms are used. According to Wikipedia, neuroprosthetics typically connect the nervous system to a device; BCIs usually connect the brain or nervous system with a computer system. Practical neuroprostheses can be linked to any part of the nervous system, including peripheral nerves, while the term BCI usually designates a narrower class of systems that interface with the central nervous system. "The terms are sometimes used interchangeably and for good reason," according to Wikipedia. "Neuroprosthetics and BCIs seek to achieve the same aims, such as restoring sight, hearing, movement, ability to communicate, and even cognitive function. Both use similar experimental methods and surgical techniques."
Convergence of Expertise
|By controlling the vertical movements of the cursor that is moving from left to right, this patient is able to select letters from groups of letters on the right side of the screen. Photo courtesy of Dr. Jonathan Wolpaw, Wadsworth Center, Albany, NY.|
Developing BCI technology brings a convergence of expertise from many collaborating institutions and individuals from such areas as motor physiology, neurosurgery, signal processing, computer programming, materials science, electrical engineering, polymer chemistry, and neurosurgery, Williams says. "We also work closely with neurosurgeons so that we can develop technologies that are clinically applicable."
"We're fortunate to be at the locus of three emerging trends," says Gerwin Schalk, PhD, a research scientist at the Wadsworth Center in Albany, New York, and associate professor of neurology at Albany Medical College. "We are gaining a better understanding of how the brain works; we are now able to better interpret these signals from the brain; we now have access to powerful computers and software that can make the needed complex calculations in real time."
BCI2000 and SIGFRIED: Versatile Use
Schalk and his colleagues have developed BCI2000, a versatile software application that supports many data acquisition devices, signal-processing algorithms, and user applications. Funded by a grant from the National Institutes of Health (NIH), BCI2000 can be downloaded at www.bci2000.org and used for free by nonprofit research and educational entities. Almost 300 laboratories and institutions around the world are now using the system, according to Schalk.
SIGFRIED (SIGnal modeling FoR Identification and Event Detection) is a signal-processing component of the BCI2000 system that provides a real-time visualization tool for evaluating complex brain signals. SIGFRIED can translate ongoing brain activity, without the comprehensive prior data collection needed by other systems, into a graphical display that can be easily understood by non-experts. SIGFRIED translates the complex brain signals obtained by BCI2000 into numbers that are mapped on a topographical display with circles marking each electrode location, Schalk explains. For instance, if the subject moves his or her hand, certain circles will grow larger, thereby indicating the locations in the brain that change their activity.
BCI2000 and SIGFRIED are currently used for two purposes, communication and diagnosis, explains Schalk. When used for communication, a paralyzed person can, for instance, use brain signals to spell words and thereby communicate with others, and to control aspects of his environment such as turning lights on and off.
The technology could have future application in prostheticlimb controls; SIGFRIED can send signals to activate a device or a visualization tool.
Functional Brain Mapping
The second major application of SIGFRIED is for diagnosing the location of important brain functions prior to brain surgery on epileptic patients. Such surgery may be considered in severe cases where medication fails to control seizures. In order to effectively plan surgeries, it is not only vital to pinpoint the epileptic focus, but also to identify the location of critical functions such as language, sensation, and movement. This process of determining critical functions is called functional brain mapping. However, all mapping techniques in use to date have important shortcomings, notes Schalk. For example, they can last several hours and induce seizures. The BCI2000/SIGFRIED system implements a novel mapping system that does not have these shortcomings, he explains. This system has been in clinical evaluation by a number of hospitals in the United States and Europe. A clinical study with ten patients at four institutions is about to be completed, Schalk adds. SIGFRIED uses complex statistical comparisons to measure the difference between brain activity recorded during a resting period and activity recorded during different tasks. This process usually takes only several minutes.
Obtaining Brain Signals
There are different methods of gathering signals from the brain. A widely used method is electroencephalography (EEG), a noninvasive method that records activity through the scalp. EEGbased BCIs are easy to use, portable, and have a low set-up cost. However, since the signals must pass through the skull, they are "blurred"-not as precise as those gained through more invasive methods. EEG-based BCIs also generally require extensive user training.
Invasive BCI devices, in which electrodes are implanted directly into the grey matter of the brain, produce the highest quality signals. However, they increase infection risk and are prone to scar-tissue buildup, causing signals to weaken or even become lost over time.
Schalk and his colleagues are exploring the possibilities of electrocorticography (ECoG), which has long been the gold standard for defining epileptogenic zones before epilepsy surgery, in BCI development.
ECoG records neural activity from a sheet of electrodes directly on the surface of a patient's brain, but not penetrating it, thus providing a less-invasive modality. Says Schalk, "I fully expect that electrocorticographic signals will provide the optimal trade-off between performance and practicality. We can access signals that have substantial fidelity without the risks associated with electrodes that are implanted within the brain. We believe that these devices will prove to be clinically practical but also extremely robust."
In epileptic patients, surgeons implant a relatively large electrode array to cover all clinically relevant areas. Insertion of this electrode array requires that a large section of skull be removed to optimize ECoG devices for the BCI. Williams is solving that problem by developing a miniaturized ECoG device that can be fed through a small hole in the skull and then unfurled to cover a larger cortical area. The electrode array, embedded in polyimide- a flexible polymer-sticks to the wet brain, thus moving as the brain moves and capturing a better signal, explains an article, "Less-Invasive Brain Interfaces," by Singer, MIT Technology Review , November 21, 2008. "It acts like Saran™ Wrap on a Jell-O mold," Williams says in the article.
When asked how the device unfurls, Williams likens it to origami, in which a complex design is folded into a small space, then unfolded. "Single sheets are molded into complex structures, drawn into a very small tube, then inserted through the cranial opening and pushed through. It then unfolds on its own."
These are just a few of the exciting BCI developments taking place, bringing more opportunities to overcome disability. Says Schalk, "We're generating powerful technologies that can now be translated-and are being translated-into applications that are changing clinical practice."
Miki Fairley is a contributing editor for The O&P EDGE and a freelance writer based in southwest Colorado. She can be contacted via e-mail: