Scientists create detailed 3-D model of human brain
With painstaking detail, scientists have created a three-dimensional virtual brain that not only maps the organ’s anatomy in unprecedented detail but also allows researchers to see how the invisible connections between cells produce the complex behaviors that make us human.
The BigBrain atlas, produced after a five-year effort, was hailed by neuroscientists as a technological tour de force that promises to speed discoveries in an increasingly important field. The work was reported in Friday’s edition of the journal Science.
“It absolutely will help us build bridges between the brain’s structure and its function,” said Dr. John Mazziotta, a UCLA neuroscientist who was not involved in the effort. “The more we understand the components of the machinery, the better position we’re in to understand how it works. It’s pretty hard to understand how a complex electronic device works if you don’t have a good wiring diagram.”
BigBrain reveals the brain’s structures with a resolution that’s 50 times better than the brain maps produced by MRI scanners. More importantly, it will make it possible for researchers, physicians and drug developers to examine the brain in a way that neither MRIs nor tissue samples on microscope slides can provide.
Looking at samples under a microscope can provide a high level of detail, but it doesn’t help researchers figure out where the samples belong in the brain or see all the cells around them.
MRI scans offer a global view of the brain and its large structures, but aren’t detailed enough to show how one type of brain cell may connect others elsewhere in the brain.
The virtual model — based on thousands of slices of an actual organ — will help researchers understand how the brain’s smallest building blocks work together to produce an array of intricate, mystifying and often amazing behaviors.
In neurosurgery suites across the world, the BigBrain atlas promises to allow more accurate location of brain tissues implicated in diseases such as depression, epilepsy, Parkinson’s disease and Alzheimer’s disease. With better guides to the cells they are looking for, surgeons can implant stimulating devices more precisely, create smaller lesions to short-circuit electrical storms in the brain, and remove tumors with less collateral damage.
It is also a key step in unifying the far-flung communities of brain scientists around a single anatomical standard — a “reference” brain by which all variations, normal and pathological, can be described, labeled and understood.
“People are pretty excited about it,” said Mazziotta, who was attending a meeting of the Organization for Human Brain Mapping in Seattle, where BigBrain was presented to scientists. In a research field awash in brain images, “what was needed was a gold standard that would have microscopic resolution. This is it.”
The brain at the heart of the project belonged to a 65-year-old woman living in Europe who suffered no neurological disease at the time of her death and willed her remains for biomedical research.
In 2008, her brain arrived — some 1,400 grams of gray tissue encased in paraffin — in the lab of German neuroanatomists Katrin Amunts and Karl Zilles at the University of Dusseldorf’s Institute for Brain Research. Assisted by an army of anatomists, the two set about carving the brain into 7,404 vertical slices and preserving them on slides. Each slice was 20 micrometers thick, roughly the width of a sheet of plastic wrap.
A year after that work began, the 7,404 slices were delivered to Dr. Alan Evans at the Montreal Neurological Institute at McGill University. Evans oversaw what he called “four years of slave labor,” in which a team of computer scientists and brain scientists worked together to line up each of those slices with perfect accuracy to create a 3-D whole.
Where delicate tissue had been ripped or cells had been ruptured, they had to be restored, Evans said. Where the “plating” of a cross-section of cells resulted in distortions or misalignment, those had to be corrected. Such exacting adjustments needed to be made on virtually every slide, often several times.
“It was a labor of love,” he said.
Arthur Toga, director of UCLA’s Laboratory of Neuro Imaging, one of the world’s largest clearinghouses of brain images, likened the process to throwing a sliced loaf of bread from a building and then reassembling the scattered pieces into a perfectly-aligned whole — as if the loaf had never been sliced in the first place.
“This project is a tour de force,” said Toga, who is moving his lab to USC in the fall. “Every step of the way posed technological challenges.”
The new map magnifies the brain atlases drawn by German neuroanatomist Korbinian Brodmann in the early 1900s and French neurosurgeon Jean Talairach in 1988 and takes them from two to three dimensions that can be accessed from anywhere in the world.
While BigBrain provides a map that approximates all human brains, it does not capture the uniqueness of each brain’s structure. The anatomy of individual brains varies in ways neuroscientists are only beginning to understand. Genes, environmental exposures, experience and disease help wire our neurons together differently.
“That is a limitation,” Toga said. “But we have to have a singular example” against which to compare the varying contours of individual brains and explore what factors helped shape them, he added.
