• eMaterials
• Faculty Spotlights
• Research Spotlights
• Links / Publications

Fall 2007

 

In this Issue:

Focus On Biomaterials

 

---

Engineering the Brain

 

Steven Schiff and Bruce Gluckman are among those who would like to find new and better ways to relieve some of the multitudinous problems that arise when the brain malfunctions. Schiff, a pediatric neurosurgeon with over 20 years of experience in the physics of dynamical diseases of the nervous system (diseases in which the structures of the organ malfunction), is working to find less destructive methods than the standard surgical procedures for treating diseases such as Parkinson’s, cerebral palsy, and epilepsy. His colleague for the past ten years in this search is experimental physicist Bruce Gluckman. Gluckman’s expertise is in the group dynamics of individual systems, with a special focus on how various parts of the nervous system interact. In 2006, the two highly regarded researchers were recruited by Penn State to form the new Penn State Center for Neural Engineering. A third faculty member, Corina Drapaca, has just been hired to add expertise in image analysis and the biomechanics of hydrocephalus.

 

“The brain is a material that is very challenging for us to deal with,” says Schiff, the Center’s director, choosing his words carefully, as well he might. Tinkering with the brain sets off alarm bells in many people. Far more than any other organ, the brain is who we are, storing our memories, maintaining our identity, expressing our emotions. Engineering the brain raises questions of our humanity, the potential of becoming more – or less – than human.

 

Schiff will approach that issue later, but for the moment he sketches the outlines of brain research. “The brain is electrical, but not like a typical circuit board. Yes, there are wires and cables and computational elements, but it is really more like a city. There is an energy supply, which is wet, and there is a lot of refuse to be collected, and that refuse, which can be simple ions, needs to be redistributed through fluid channels. It’s like that book that shows all those things that are hidden under the streets and buildings of New York City that make it run.”

 

Bruce Gluckman, the engineer to Schiff’s clinician, adds another layer of metaphor. “I look at it this way – if you consider a business in a big office complex, if the plumbing backs up, everyone leaves the building and no work gets done. The success of the business depends on the plumbers and the garbage collectors. The whole support structure has to be designed well. When parts of that break down, the whole business fails.” In the brain, the support system can fail in subtle ways. It’s not all about measuring the electrical activity connected with computation, Gluckman says, it’s also about the support structure that keeps the brain functioning correctly – the metabolism.

 

Materials for Neural Engineering

The new Center for Neural Engineering will be a bridge between the clinical researchers at the Penn State Hershey School of Medicine and the engineering and materials disciplines at University Park. Neural engineering relies on materials science as well as computation and electrical engineering to attempt to restore brain functions. There is the potential that cognitive disorders such as depression, obsessive/compulsive disorder, and other things that plague people might be related to rhythms in the brain that could be modulated with systems that can sense, then compute, and then actuate stimuli which could alter how the brain is functioning in ways that simply bathing the brain in drugs can’t, Schiff says.

 

For instance, Gluckman and Schiff have learned that low frequency electrical stimulation can modulate parts of the brain in order to turn off the seizures and symptoms of Parkinson’s disease or epilepsy. However, to truly take advantage of this knowledge, they need to understand where the electrical signals come from and how they affect the brain. That has proven to be a problem. The interpretation of EEG (electroencephalography) results, which record the electrical activity of the brain using electrodes placed on the scalp or within the brain, are no clearer now than when Schiff was in medical school. “In the last ten years we’ve thrown an enormous amount of computer cycles at trying to understand the electrical activity of the brain through better signal processing, and we’ve learned, what would you say, Bruce – nothing?” Schiff says with a shrug.

 

Without a good physical model of how the brain works, the interpretation of electrical activity remains guesswork based on the superficial resemblance of patterns that show up on an EEG. “You spend a year or two in a dark room,” Schiff says of the training physicians receive on the electroencephalograph. “When you see the same patterns as on a medical diagnosis, you’re allowed to graduate.”

 

Gluckman and Schiff came to Penn State, at least in part, because they felt that here was a place ideally situated to solving the mystery of how the brain works, especially those subterranean systems that provide the brain’s infrastructure. So far, no one has designed the sophisticated sensing devices to show what’s going on in the brain. They don’t know where to place the sensors or how large they should be. Do they all have to be in the head or can they pick up enough information outside the head and avoid the need for surgery? With what Schiff calls Penn State’s “enormous sophistication” in testing the electrical and magnetic properties of materials, “if they knew how to apply that knowledge to nervous systems, then, who knows? In a perfect world, we could be the two people who could help them make that bridge.” (In a new Materials/Life Science complex, expected to open in the next three years, the Center for Neural Engineering’s researchers will work alongside materials scientists and engineers expert in nanotechnology and microfabrication.)

 

It Isn’t Rocket Science

Neural engineering is the techie side of neuroscience, says Gluckman. Its earliest big success was the development of the electroencephalogram, which has been in wide use for over half a century and gave neuroscientists at least a blurry blueprint of the electrical activity of the brain.

 

More recently the major advances have occurred in devices and prosthetics, especially the cochlear implant for hearing loss, first developed in 1969 and now worn by over 100,000 users worldwide. Other active projects involve other sensory inputs, such as retinal implants, and artificial limb implants. Recently, a great deal of effort has gone into developing the braincomputer interface, a closed loop system in which signals feed back and forth from the brain to the computer, in effect training both.

 

“It has been five years since animal studies showed monkeys controlling robotic arms or cursors on a screen through measurements out of the motor cortex,” says Gluckman. Similar experiments have been performed with human volunteers, with electrodes implanted that allow some control of computers. This has also been done for “locked-in” individuals who through brain trauma or disease have completely lost the ability to move or communicate. In those cases, the computer-brain interface is made in a noninvasive fashion that uses EEG-type recordings.

 

“There are also successes in neural engineering for neurological problems such as Parkinson’s disease and epilepsy, which I have been studying for the past ten years, and Steve for more like 20 years,” says Gluckman. Schiff adds, “Neural engineering has been around for a long time, but the applications were spotty. There have been a few examples of very successful implementations, but it has been in the last 10 years that we have been able to blend what we know about materials engineering, electronic engineering, and computer engineering with working with a soft and wet three-pound computational organ.”

 

Despite these recent successes, sensing and control systems for the brain are still unsophisticated. The cruise control system in your car is probably substantially more sophisticated than any neural control system that’s been designed so far. “It’s not rocket science,” Schiff says, speaking quite literally. “Back in the sixties, rocket science reached a level of engineering theory that allowed it to do all those things we know about from the space program, sending rockets throughout the solar system and landing humans on the moon. We’ve never applied a whit of that knowledge to working with neural control systems of the brain.”

 

The purpose of sensing and control devices is to treat diseases in which the dynamics of the brain are malfunctioning. In such cases the brain is like a turbine that’s beginning to wobble, and scientists such as Schiff and Gluckman use control pulses to try to stabilize its frequency. “We’re beginning to understand that associated with cognitive disorders of the brain there seem to be changes in rhythmicity” Schiff says. “Broadly speaking, how do you sense that a nervous system is not functioning properly, that its rhythms are wrong, and how can you readjust those rhythms by stimulation? The obvious case that comes to mind is Parkinson’s disease.”

 

Into the Lab

The Neural Engineering Center is made up of three connected laboratories on the floor with Gluckman’s and Schiff’s offices in the Earth-Engineering Sciences Building, one of the cluster of new research buildings on the west end of campus. The first lab is a small shielded room with hardware and software devoted to computer/ brain interface experiments. On a table on one side of the lab, a toy vehicle is attached by wires to a computer. Here students and postdocs wearing electrode-studded skull caps attempt to control the movements of the truck using only the electrical signals from their brain. “In the future,” says Schiff, “we are going to have a course where our students play neural ping pong as one of the laboratory exercises. We want to guide students into projects that lead into the unknown.” This room serves to prototype coursework and equipment for a future Brain-Machine Interface teaching laboratory at Penn State – space for it is already reserved in the future Materials/Life Sciences II building. Penn State will be one of the first sites in the world to offer such a course for its engineering students.

 

In the second lab, two students are studying small slices of brain under microscopes. Gluckman explains the purpose of brain slice research. “About a decade ago, Steve and I rediscovered that if instead of using pulses like in most deep brain stimulation, we used some very low frequency electrical fields or electrical current, meaning a constant electrical field that lasts longer than the firing of a single neuron, then we could shift the set point of the cell so that slightly more or less input is required to fire the next signal. This modulates the response of that neuron to its input.” For instance, they can make the neurons less responsive to the electrical firestorm of a seizure.

 

After years of brain slice experiments, the two scientists discovered they were able to simultaneously record the activity of the system as they were modulating the response – creating, in effect, a closed loop. The feedback system allowed them to control seizure-like activity in the brain slices, and eventually to create control devices to interact with seizures in implanted animals.

 

In the third lab, the researchers translate the knowledge gained in the brain slice experiments to understanding the complexities of electrical activity in living subjects. Here Gluckman builds small implantable electronic circuits to record the firing of single neurons under low frequency. In one of the thumbnail-size devices, he has integrated the kind of commercial accelerometer that can be found in auto airbags into a circuit designed to measure the tilt of a lab animal’s head, which helps determine behavior (sleeping, drinking) when combined with the EEG. One goal is to use such data to predict a seizure before it happens. Also in this lab, with a grant from the Grace Woodward endowment at Penn State, they are developing an X-ray CT scan analysis algorithm to diagnose temporal lobe epilepsy in children in sub- Saharan Africa that would normally require costly MRI equipment.

 

The Implications are Huge

Nanotechnology and microengineering are highly advanced research areas at Penn State that can speed the progress of neural engineering. Gluckman predicts that the techniques of nanotechnology will help them develop the elusive model of the brain that is currently lacking. “We think that the whole science that is developing around creating different surfaces with different textures and structures and materials is one that is extremely important for neural engineering,” Gluckman says. “We are deeply involved with sensing, using nanosensors to study analytes in microdomains of the brain in both space and time. What we mean by that is you need to know the level of metabolites that change the excitability of electrical components of the brain. By using various technologies to do sensing on very small scales, and knowing the anatomical structure, we can see how materials are transported and build computational models to understand how that alters the stability in parts of the brain.”

 

Even without a clear understanding of how the brain works, dramatic possibilities are opening up, says Schiff. “There have been very high profile papers coming out in the scientific press in the last couple of years showing you can modify memory in people by applying electrical current to the brain. A recent paper showed you can disrupt cancer cells with modestly high frequency fields that are focused properly. Because it is not well understood why this works, it is important to get better models. It’s something we feel we can provide here at Penn State.”

 

Functional magnetic resonance imaging (fMRI) is one of the important new tools for neural research. Rather than measuring electrical activity (EEG), the fMRI measures blood flow, indicating areas of the brain that are presumably involved with mental and physical activities. Penn State is about to get a new magnetic resonance imaging tool on campus for human use. The ability to make magnetic and electrical fields is enormously expanded by that addition, Schiff says. One example of the utility of fMRI is a paper that appeared in the journal Neuron about a year and a half ago, showing that a very small target very deep in the brain when stimulated for most of the day could substantially help the majority of the small handful of patients they tested with serious depression. They picked the target for stimulation based on functional magnetic resonance imaging of people who were or were not depressed, and determined from such images who was a good candidate to stimulate. Amazingly, it worked, Schiff says. We can look for an explosion of such technology following on the heels of this study.

 

Schiff argues for caution in the rush to harvest the benefits of neural engineering. “We’ve made a big push here to be very careful to study the damaging effects of putting something into the brain, and to be very careful about how even our electrical stimulation can damage it. We want the things we make here at Penn State to be safe for the patients that we treat. All medical decisionmaking balances risks and benefits. We’re very focused here on paying attention to both. I hope over time we can collaborate with the experts on campus to do things none of us could ever do on our own.”

 

Contacts:

• Steven J. Schiff, MD, PhD, is Brush Chair Professor of Engineering and director of the Penn State Center for Neural Engineering: Validate to view address - Send Email via form
• Bruce J. Gluckman, PhD, associate professor of engineering science and mechanics, is Associate Director of the Penn State Center for Neural Engineering: Validate to view address - Send Email via form

 

For more information on the Center, visit: http://www.esm.psu.edu/wiki/research:cne:start

---

Download the PDF (4.29MB) version.

Sign-up online to subscribe to Focus On Materials.