The Human Brain, Ready for Its Close-up

In his book, The Mind’s Eye, neuroscientist Oliver Sacks describes the brain as “a vastly complicated orchestra with thousands of instruments, an orchestra that conducts itself, with an ever-changing score and repertoire.”

For decades scientists have studied the “orchestra” through neuroimaging technologies, such as MRI, and through large collaborations, such as President Obama’s Brain Initiative, to better understand the brain’s anatomy and afflictions. Despite these efforts, however, we still know very little about that three-pound organ in our heads. For instance, researchers have difficulty pinpointing exactly where various brain diseases arise, or where regions such as the frontal cortex begin or ends.

A new project recently announced in the journal Science, called BigBrain, could give us the close-up we need to solve some of these cranial conundrums. Developed as a part of the European Brain Project, the BigBrain is the most detailed three-dimensional brain model to date. Its ultra-high resolution is the closest scientists have gotten to seeing the brain on a cellular level.

To create the model, German and Canadian researchers first used a ham slicer-like device called a microtome to divide the healthy brain of a deceased 65-year-old woman into 7,404 sections. Each slice, as thin as a Saran Wrap sheet, has a resolution of 20 micrometers, or 50 times the resolution of an MRI. The researchers then stained and scanned these sections into a computer, which assembled them into a comprehensive atlas that homes in on various brain components, including the cerebellum, and reveals structures such as the cell bodies of neurons. The sections account for more than a terabyte of data, which is more than 125,000 times the data gathered from an MRI, according to study author Katrin Amunts, a neuroanatomist at the Research Centre Jülich in Germany.

This new atlas is a vast improvement over the world’s first comprehensive brain map, introduced in 1909 by German anatomist Korbinian Brodmann. His map, known as the Brodmann Areas, divided the brain into 52 regions. “In these traditional brain maps, there was the assumption that ‘Region 1’ talked to ‘Region 8,’ but their interactions could not be explained,” said Alan Evans of McGill University’s Montreal Neurological Institute. “We’re looking to reproduce Brodmann’s work so we can see, in finer detail, what the connections and relationships are between these regions.”

Evans expects that BigBrain may lead to major improvements in medicine. A surgeon, for example, could potentially pull up a section of the atlas to guide her incisions and avoid damaging any critical areas. Physicians will also be able to compare BigBrain’s high-resolution images with images of diseased brains, says Stuart Lipton, a neurologist at the Sanford-Burnham Medical Research Institute in California. They can then better understand what connections are disrupted in a brain with a condition like Alzheimer’s or autism.

Other scientists working on brain-mapping research look forward to using BigBrain to inform their work as well. Daniel Marcus, director of the Neuroinformatics Research Group at Washington University, contributes information to the Human Connectome Project, which is devoted to enhancing knowledge of brain diseases and mental disorders. He says the BigBrain’s ultra-high resolution would be useful in checking the accuracy of the data he receives from MRI studies.

Marcus notes, however, that when studying certain diseases such as dementia, which changes the brain over time, it can make more sense to use MRI instead of a high-resolution atlas based on one person’s brain. Plus, “it’s a postmortem brain,” he says, “so we can’t really see how it functions and how diseases develop over time. The MRI, meanwhile, is noninvasive and can track changes.”

The BigBrain will also complement another mapping effort called the Allen Human Brain Atlas, which is more focused on identifying gene expression in the brain for the Human Genome Project. The Allen Institute, where the project is based, uses postmortem brains as well, which are divided into sections that run at higher resolutions than MRI images, though not as high as BigBrain’s.

“BigBrain will allow us to learn more about the brain’s construction,“ says Mike Hawrylycz from the Allen Institute. “The model will have important applications once the boundaries between regions in the cortex and the subcortex are defined and labeled.”

Researchers working on the BigBrain project plan to slice up more brains to build a better sense for what a healthy one looks like on a microscopic level. Eventually, they want to achieve an even higher resolution of two microns. While the tools to slice sections that thin are available, says Evans, the next step is finding a computer that can hold all of the data associated with such detail.