Studying the human connectome to increase our understanding of epilepsy

// Biomedical Engineering

Tags: biomedical engineering, Elizabeth Meyerand, epilepsy, research

Photo of Beth Meyerand in her office

Beth Meyerand studies brain maps of epilepsy patients and healthy controls to identify differences in connection patterns that may contribute to the disease. Photo by Stephanie Precourt, College of Engineering.

Share this story:

Francis Collins, the director of the National Institutes of Health, has called the human connectome “the symphony in our brain,” played by electric circuit connections between its distinct functional regions.

Neurologic diseases can distort that intricate symphony—yet researchers still have a poor understanding of how that happens, prompting Collins to launch the NIH Human Connectome Project in 2009.

Its first goal was to map normal connections, or communication patterns, among the nerve cells in different areas of the brain. Next, NIH invited research groups around the country to submit proposals for comparing those normal patterns to malfunctioning connections caused by different brain diseases.

Researchers at the University of Wisconsin-Madison and the Medical College of Wisconsin in Milwaukee were successful in securing funding for not one, but two of those collaborative proposals: one for Alzheimer’s disease and one for epilepsy. With software and technology development playing an especially important role in this field, the epilepsy project’s co-principal investigator is engineering professor Elizabeth Meyerand.

“The success of our proposal was due, in part, to the paradigm shift we initiated in the epilepsy community in the early 2000s,” says Meyerand, who is a professor of biomedical engineering and medical physics at UW-Madison. “Until Bruce Hermann, a neurology professor at the UW School of Medicine and Public Health, and I demonstrated that changes in the brain’s white matter play an important role in epilepsy, it was thought to be purely a grey matter disorder.”

Meyerand compares the brain’s grey matter to cities on a map and its white matter to the roads that connect the cities. She describes the white matter changes she and Hermann detected as bumpy roads with chipping asphalt in epilepsy brains, compared to nice and smooth surfaces in normal brains. The researchers found these changes with diffusion tensor imaging—a new technique at the time—which is a special form of magnetic resonance imaging (MRI) that provides pictures of the brain’s white matter tracts.

In the NIH-funded epilepsy connectome project, which is completing its second year in fall 2017 and will continue through 2019, Meyerand and colleague Jeffrey Binder, a neurology professor at the Medical College of Wisconsin, are now using MRI and other techniques to map—in unprecedented detail—the brains of 200 healthy adults and 200 adults with temporal lobe epilepsy, the most common form of the disease that affects at least 50 percent of all patients.

Meyerand and Binder are especially motivated by the plight of almost 900,000 American epilepsy patients who don’t respond to anticonvulsant medications. Those 35 percent, out of 2.5 million diagnosed with the disorder, continue to have regular, sometimes daily, seizures that severely limit their ability to lead a normal life. The next level of treatment for these patients is brain surgery, with a moderate success rate of 40 to 50 percent. Both prescription drugs and surgery focus entirely on the brain’s grey matter.

“Since the existing treatments don’t work for a large percentage of patients, we clearly don’t understand the disease well enough yet,” says Meyerand. “That’s why this research is so important.”

For the epilepsy connectome project—the only one of its kind in the world—study participants complete three kinds of MRI scans; patients also provide extensive neuropsychiatric interview data. In addition, participants who are seen in, or are willing to travel to, Milwaukee undergo magnetoencephalography (MEG) scans, which require specialized equipment not available in Madison.

“The MRI scans provide detailed images of the anatomy, 3D size and connections between different brain regions,” Meyerand explains. “As a complementary method, the MEG scans give us real-time measurements of brain activity. Both types of scans are taken while people are resting and relaxing, and also while they perform certain cognitive tasks.”

Since every action we perform, and every thought we have, generates a moving electrical current in our brain, which in turn creates a small magnetic field, MEG is a more direct measurement of brain activity than MRI.

Resting state scans measure the brain’s “housekeeping” functions, which Meyerand compares to construction workers just checking in to let each other know they are on site, working on their respective part of the project. Functional scans measure connections between those brain regions that are activated during certain tasks, such as finding all possible opposites of the word “good,” memorizing a list of words, or recognizing faces.

With the scans in hand, Meyerand’s team uses computer models to produce color-coded brain maps. Next, the researchers identify potentially meaningful differences in the brain connectome of epilepsy patients and healthy controls with statistical methods. They also examine whether a patient’s seizure frequency or response to treatment can be predicted by their baseline MRI or MEG scan results.

The extensive information about the healthy brain’s connectome will be a tremendous resource for multiple other research studies going forward. In addition, since both the patient and control data—in de-identified and standardized form—are stored in an NIH-funded data repository, they can be freely downloaded and analyzed by many other researchers, greatly speeding up the rate of scientific discovery, according to Meyerand.

“This project is a perfect example for the kind of big and impactful science that can only be done by a large and highly interdisciplinary research group, in this case representing biomedical engineering, neurology, medical physics, radiology, psychiatry and statistics,” Meyerand says. “By targeting epilepsy treatment to each person’s unique connectome information—an example of personalized medicine—I am optimistic that patients will eventually experience much better long-term seizure control than we are able to achieve today.”

Author: Silke Schmidt