A cadre of University of Minnesota researchers is coleading the Human Connectome Project—the most ambitious brain-imaging study ever conducted
Imagine a road map connecting every one of Earth’s 7.1 billion people — and showing how each of those people is connected to the 300 or so people he or she knows. Now imagine 11 more identical maps, crumple them all up, stuff them into a cantaloupe, and try to read them. Now you’ll begin to have an idea of the complexity of the “human connectome,” as researchers refer to a comprehensive map of neural connections in the brain.
The average human brain has about 85 billion neurons, each of which has thousands of connections to other neurons. The whole intricate network is compressed into roughly three pounds of delicate, living tissue. It is the most complex and least understood organ in the human body.
The Human Connectome Project, or HCP, has brought the combined brainpower of 36 investigators — including six from the University of Minnesota — from 11 different institutions to bear on the challenges of understanding the human brain. Now just past the midpoint, the five-year, $30 million research effort is mapping the vast network of neurons and the trillions of interconnections that compose this elaborate organ.
“We’re using some of the most advanced magnetic resonance imaging (MRI), data acquisition, and data analysis tools ever used to study the brain,” explains Kamil Ugurbil, Ph.D., HCP project coprincipal investigator, and founding director of the University’s Center for Magnetic Resonance Research (CMRR).
Ugurbil, a McKnight Presidential Endowed Chair, is spearheading the research efforts of the Minnesota contingent — which is focused on optimizing the MRI hardware and the data acquisition processes and software. He is a pioneer in high-field MRI — which allows scientists to “see” what’s going on inside a living human brain — and helped to develop the world’s first 7T scanner that provided preliminary data for this grant. (The T stands for “Tesla,” a standard unit of measurement for magnetic fields; 1.5T MRI scanners are commonly used for imaging in hospitals and clinics.)
A project defined
In 2010, the National Institutes of Health awarded almost $40 million in grants to support brain mapping and imaging research efforts by two consortia. The largest grant — $30 million over five years — was awarded to a consortium led by Washington University in St. Louis and the University of Minnesota.
The WU-Minn consortium was tasked with optimizing MRI scanners, related technologies, and data acquisition methods to capture the most advanced images (maps) ever of the brain’s anatomical wiring (using diffusion-weighted MRI) and its activity (using functional MRI or fMRI), both in resting state and while the subject is performing various tasks.
In total, the WU-Minn connectome project will acquire high-resolution MRI brain scans from 1,200 healthy adults. This group will include 300 sets of four siblings — each set consisting of a pair of twins and two non-twin siblings. In addition to brain scans, the project will collect genetic information and behavioral data for each subject. Washington University is conducting the majority of the scans, but the University of Minnesota will conduct even higher-resolution scans — both 7T and 10.5T — of a subset of the larger group. The researchers believe that by comparing these different scans of the same individuals, they will be able to learn more about how various parts of the brain interact.
(A second consortium, led by Massachusetts General Hospital/Harvard University and UCLA, received an $8.5 million grant to perform similar studies using diffusion MRI technology for imaging the brain’s structural connections.)
Both research groups will make their data freely available to neuroscientists and other researchers around the world. The WU-Minn consortium has begun releasing data from the scans acquired to date.
Founded in 1991, the University of Minnesota’s CMRR has established itself as a pioneer in medical imaging, thanks in part to support from the W. M. Keck Foundation. Today it is home to some of the world’s strongest magnets (including a 16T magnet for animal research and, arriving soon, a 10.5T for humans), and its faculty is widely recognized for making MRI scanners bigger, faster, and capable of producing higher-resolution images.
“The University of Minnesota team has played a critical role in creating the improved methods of data acquisition,” says HCP coprincipal investigator David Van Essen, Ph.D., Alumni Endowed Professor of Neurobiology at Washington University. This includes the development of advanced pulse sequences for a customized 3T scanner that was initially housed at the CMRR, then sent to Washington University, where it now serves as the workhorse scanner for the target number of 1,200 healthy adults. In addition, the U of M team is using a similar approach to refining the data for a 7T scanner housed at CMRR, which will be used to study 200 of the same participants starting this fall.
Essa Yacoub, Ph.D., a CMRR associate professor and HCP investigator, helped to optimize the technologies used in both the 3T and 7T scanners.
“My expertise is in pulse sequences — essentially what you program the MRI machine to do in terms of acquiring data,” says Yacoub, emphasizing that his efforts are supported by a whole team of engineers, technicians, and other research associates.
The U of M team has developed multiband pulse sequences that enable scanners to capture imaging data or pictures from up to eight “slices” of the brain at one time.
“A standard clinical MRI scan takes a picture of one portion of the brain at a time,” says Yacoub, explaining that acquiring multiple pictures permits faster scanning, which in turn allows for improved image clarity and much better data overall because of the dramatic increase in efficiency. Faster data acquisition also makes it easier to follow dynamic processes in the brain.
Although HCP investigators have already racked up some very significant achievements, in some ways, their work is just the lever that starts the ball rolling. Most likely, it will be those who use the HCP data — and the new HCP-generated technology — who will catapult understanding of the human brain to the next level.
Two University of Minnesota researchers, for example, are in the midst of research projects that have benefited from HCP advances in pulse sequencing.
Christophe Lenglet, Ph.D., an assistant professor in the U of M Department of Radiology and an HCP investigator, helped to evaluate the multiband sequence technology that sped up data acquisition. Now, he’s using that technology, with collaborators Gülin Öz, Ph.D., Isabelle Searcy, Ph.D., and Khalaf Bushara, M.D., in research projects focusing on ataxia.
“Multiband sequencing enables us to acquire a lot of data in a short amount of time,” he says. “We are looking for changes in the brain wiring, which requires hundreds of 3-D images to be acquired and then analyzed with advanced computer software. Preliminary results have shown some alterations in regions of the brain that we expected — and some that we didn’t.”
Psychiatry professor Kelvin Lim, M.D., is also making use of the higher-speed imaging capabilities.
“We’re working on a study using MRI to map brain connectivity,” says Lim. “The more data we can gather the better, but our subjects cannot stay still in the scanner for a long period of time.
“By using the multiband sequences developed in the HCP, we can collect data eight times faster than before. It’s like magic.”
The HCP also recently released the first wave of data generated by the project. Neuroscientists and other researchers may apply online at the HCP website for free access to it.
Thomas Yeo, Ph.D., a research fellow at Duke-NUS Graduate Medical School in Singapore, has downloaded the data — and he’s excited about the doors it is opening for him.
“I have worked with the data from HCP and obtained promising results regarding the topography of interactions among human brain networks,” Yeo says. “I suspect that these interactions are crucial to understanding the neural basis of individual differences and what makes each of us unique.”
As more researchers use the growing HCP database, the prospects for significant breakthroughs in neuroscience multiply.
“This is a very exciting project,” says Ugurbil. “We’re laying the groundwork for other scientists to one day begin to identify effective treatments and perhaps even cures for many very troublesome brain disorders and diseases.”
Freelance writer Chuck Benda is a former editor of the University of Minnesota alumni magazine and has written dozens of articles about the University over the past 30 years.
To support the leading-edge work under way at the Center for Magnetic Resonance Research, contact Patricia Porter at 612-626-6703 or firstname.lastname@example.org.