the florida engineer

The research featured here, besides being complicated, is produced by the most recent additions to the Gator Engineering faculty — all hired in the last eight years. And the research runs the gamut from cancer treatment to tracking ingested medication to curbing traffic congestion.

Pramod Khargonekar made it a point to infuse the Gator Engineering faculty with some of the best, most promising junior faculty members in the country.

Vaccines with a Kick

Biomedical engineer Ben Keselowsky is collaborating with UF’s pathology and urology departments to develop vaccines for Type 1 diabetes and prostate cancer.

He is working on a high-throughput approach to create, test and optimize vaccine combinations that include biomaterial particles. “These diseases have disrupted immune responses, so you basically want to design a vaccine that will tune to the body’s immune responses,” he said.

Keselowsky is working on designing the best biomaterials for these vaccines. The biomaterials are targeted at dendritic cells. These are ideal targets because they process foreign invaders and other materials in the body and present them to the other immune cells as a way of telling those cells what materials to look out for and respond to.

“They’re the key orchestrator of immune response, balancing immune response to foreign invaders and promoting tolerance to self-antigens,” Keselowsky said.

The prostate cancer vaccine’s goal is to cause the dendritic cells to stimulate immunogenic responses, causing the other immune cells to attack and destroy tumor cells. The goal for the diabetes vaccine is to stimulate a tolerogenic response, which will cause the immune cells to stop attacking insulin-producing pancreatic cells. Both vaccines are biomaterial particle-based, so they can provide a biodegradable, time-released depot of antigen and immuno-modulatory molecules. These particles can also incorporate dendritic cell-targeting molecules as well as chemoattractants and differentiation factors.

Keselowsky said his team developed a novel microarray platform consisting of arrayed islands of dendritic cells co-localized with adsorbed particles.

“This allows us to probe dendritic cell responses to a thousand different vaccineparticle formulations simultaneously on a single chip, enabling the discovery of unforeseen interactions between vaccine components,” he said.

Radiation Detection

Glenn Sjoden’s research on detecting special nuclear materials enables him to analyze and optimize neutron and gamma ray transport for detection, nuclear power or forensics in almost any scenario. Most of his research deals with three-dimensional transport theory, which uses parallel codes and highperformance parallel computers to estimate the levels of radiation and its effects as it transports and scatters in a system.

“Knowing where the radiation goes is priceless,” said Sjoden, UF associate professor of nuclear engineering.

With co-author and nuclear engineering chair Alireza Haghighat, he has written some of the codes used in transport theory applications, including one code called PENTRAN recently shown to solve a system of 5 billion unknowns on thousands of processors.

“Right now, it’s the only code in the world tested to that level,” he said.

Sjoden and the UF Transport Theory Group worked with IBM and Tanguy Courau, a visiting scientist from Electricite’ de France the past year to analyze a whole reactor core with PENTRAN.

Sjoden said he and his graduate assistants at the time — Kevin Manalo, Tom Plower and Mireille Rowe — adapted PENTRAN to analyze fuel burn-up in a nuclear reactor with a post processing code called PENBURN. They used PENTRAN with PENBURN because the code solves unknowns in angle, energy and space over a three-dimensional mesh.

Robots, Satellites, Muscles, Oh My!

Warren Dixon could help stroke patients gain more precise control of their muscles and motion during rehabilitation while experiencing less fatigue. He’s also using the same technology to improve satellite and robot function.

Dixon, an associate professor of mechanical and aerospace engineering, is developing new mathematical formulas to predict the behavior of nonlinear systems, which can lead to better control. Since nonlinear systems include anything from robots to satellites to the human body, Dixon’s research has numerous applications, all falling under the same mathematical umbrella. “We have this philosophy that the more information you have about the way a system is going to behave, the more sense you have to be able to do what you want to do,” he said.

The common theme all of these applications is Lyapunov theory, which is based on “if a function is lower bounded and always decreasing, then the function will remain bounded and converge to the lower limit,” Dixon explained. He said his research group is building on Lyapunov methods to construct new control designs and stability analyses.

One of Dixon’s most intriguing projects deals with muscle stimulation and could help stroke patients in rehabilitation.

Muscle activity is one of the most nonlinear and uncertain systems because it varies depending on the person and the circumstance, Dixon said. Nonetheless, his group tries to develop a Lyapunov-based mathematical formula that takes into account factors such as muscle fatigue, muscle fiber type and pH levels and tries to predict the best level of stimulation required to move the muscle a certain amount.

Making It Flow

Using a vehicle that records data such as how a driver changes lanes and how much distance a person keeps between him and other cars, allows researchers to categorize drivers as aggressive or conservative, explained civil and coastal engineering professor Lily Elefteriadou.

The researchers then incorporate driver behavior into the algorithms at the core of traffic simulation software. So if a city knows which types of drivers are most common in its population, then it can create a more realistic simulation to figure out the optimal highway design.

“The presence of very aggressive or very conservative drivers has an impact,” said Elefteriadou, director of the Center for Multimodal Solutions for Congestion Mitigation. For example, cities might want to adjust the number of lanes on a highway because aggressive drivers tend to quickly move over to the left lane after entering the freeway, thus causing all of the surrounding cars to accelerate, she said.

The CMS was established in May 2007 and is federally funded. It is one of only a handful of such centers in the U.S. and is affiliated with the Transportation Research Center, which has been at UF since 1972. Elefteriadou is also the director of the TRC.

The traffic simulation software developed at the CMS is distributed through McTrans, another affiliate of the TRC.

Elefteriadou said that although several universities develop and use their own simulation software,

UF’s software is the most widely distributed for commercial purposes.

Disrupting Cancer

After a tumor is removed from a cancer patient’s esophagus, the patient would likely receive a stent to keep the now-weakened esophagus walls from collapsing. But a common problem with an esophagus stent is cancer cells adhere to the stent surface and then multiply on it, causing more blockages, said chemical engineer Tanmay Lele.

This process requires the generation of intracellular mechanical forces that are exerted on the stent surface. To keep the cells from adhering to the stents, Lele tries to stop the cells from exerting those forces in the first place by creating nanostructures that disrupt the molecular-level processes happening on the cell’s surface.

The nanostructures specifically disrupt the clustering of transmembrane receptors called integrins, which recognize and bind to adsorbed proteins on the stents. The integrins need to cluster to transmit force from the intracellular actomyosin cytoskeleton to the stent surface.

Lele hopes to disrupt tumor-cell attachment to the stent surface by spacing out the nanostructures to prevent integrin clustering. Initial experiments show strong potential for this approach, he said.

The American Heart Association has funded a great deal of Lele’s research because altered mechanical forces are also responsible for artery blockages, so understanding how cells respond to mechanical forces is key to understanding heart disease, he said.

Size Does Matter

An integrated circuits design group is developing lowpower microsystems to sense and process brain signals using various steps of algorithms. The microsystems then wirelessly transmit those processed signals to an external device that controls movement, explained electrical and computer engineering assistant professor Rizwan Bashirullah.

Since these microsystems would be implantable electronic devices, they must be small and low-power but highly functional; they have to be able to sense the low-noise and highly parallel brain signals while using a minimum amount of power so the device does not overheat, said Bashirullah, an assistant professor of electrical and computer engineering.

As a self-described “hardware guy,” Bashirullah said he focuses on developing algorithms that can work within these constraints.

“Not every algorithm is hardware- or energy-efficient, so one of the things that we’re trying to do is basically create algorithms that are very hardware-efficient in terms of the amount of space [and] the amount of computational energy required,” he said.

He’s also working on a separate low-power microsystem that could save the pharmaceutical industry billions of dollars each year, which could translate into lower health care and drug costs.

Bashirullah is developing a biocompatible electronic microchip that can be placed on a pill capsule and monitored from outside the body to track the pill’s trajectory.

“It’s an electronic tag or time stamp associated with the patient ingesting the pill,” he said.

As of now, it’s difficult to track whether patients in a clinical trial are actually taking the prescribed drug, which makes it hard to infer the accuracy of the results, he said.

Like the microsystems that process brain signals, this device is small, wireless and low-power, he said.