(click to enlarge)
A portable wireless system records neuronal modulation data downloaded from the implantable brain chip. Graphic reprinted with permission from IEEE, 2007.
With the ability to process brain signals and stimulate neurons, a neuroprosthetic chip could have the power to prevent seizures and control artificial limbs. Researchers at the University of Florida (UF; Gainesville) have the components needed to create the brain chip. Now they must successfully integrate the technologies into one package.
Initial applications of the device focus on treating paralysis and epilepsy. The chip will be able to sense a problem or intent from a person's brain, decode the signal, and then process it to deliver therapy through the device's interface.
“The brain chip is a means to bypass damaged tissue in the nervous system,” says Justin Sanchez, PhD, assistant professor of pediatric neurology, neuroscience, and biomedical engineering at UF. “We can use engineering principles to help resolve some of the functional deficits that may be in the nervous system.”
The researchers have already developed an algorithm and tested new types of electrodes. These will interface with the brain to provide signal sensing. Data that are recorded on the chip can be downloaded wirelessly, recorded onto a separate system, and then reviewed by the user at a later time.
“We've developed a new signal representation that enables the fully implantable device to send out many channels of neuronal modulation synchronously and wirelessly,” says Sanchez. “We have a pulse-based representation of neuronal firing that we can use to overcome some of the power and bandwidth issues.” The researchers also developed new signal and coating techniques to obtain more data from the implant.
The chip will be about 10–12 mm in diameter and 1–2 mm thick. It will be implanted between the top of the skull and the bottom of the scalp.
Justin Sanchez from the University of Florida says the chip could bypass
damaged cells in the nervous system.
The main design challenge will be to combine the electrodes, amplification, signal-processing, and wireless technologies into a small, portable device that is rechargeable or can run on a primary battery for a long period of time. The researchers also face power constraints associated with wireless data transmission and sending multiple channels of data in real time.
To use the brain chip to treat paralysis or epilepsy, the device must be able to decode and process the neural representation to deliver therapy. To do this, the researchers must improve the device's signal analysis. This will help the implant determine a patient's intent to move a prosthetic limb or control a wheelchair, for example. When attempting to prevent epileptic seizures, the device must identify abnormally firing neurons in a reasonable time frame to be able to disrupt the irregular neural activity.
The National Institutes of Health gave the researchers a $2.5 million grant for their work. Sanchez is working closely with colleagues at UF's College of Engineering on the project. Jose Principe is involved in the signal-processing aspect; John Harris is handling analog circuit design; Rizwan Bashirullah is responsible for the wireless system; and Toshikazu Nishida is involved with the microelectromechanical system and the fabrication of new electrodes.
The next step for them is to incorporate the individual technologies into a complete system, test the components together in vivo, and show proof of concept. A prototype could be ready for human testing within a few years.