查看原文

The digital three-dimensional model called "BigBrain" was produced from the thousands of sections made from the brain of a 65-year-old woman. Its resolution is finer than a human hair, so it can reveal clusters of brain.

A 65-year-old woman’s brain was cut into 7,400 slices to create the most detailed three-dimensional atlas of the human brain ever made, bringing researchers one step closer to reverse-engineering the brain’s convoluted circuitry.

Brain atlases are essential reference tools for researchers and physicians, to determine which areas are “lighting up” during a task or thought process, or during image-guided surgery. The better the atlas resolution, the better doctors can target ever-smaller parts of the brain and their individual function.

The atlas creators, who are from Canada and Germany, have made the ultrahigh-resolution model — 50 times more detailed than a typical scan — publicly available in a free online format. The authors also published their work in the journal Science on Thursday.

The atlas, called BigBrain, offers a common basis for open, worldwide scientific discussion on the brain, said author Karl Zilles of the Heinrich Heine University Düsseldorf.

Zilles pointed to a novel treatment for Parkinson’s disease called deep brain stimulation, where electrical impulses are sent through electrodes implanted into specific points in the brain. He said BigBrain may open the doors for more accurate localization of electrode placement and thus render treatment more effective.

After staining and digitizing the thousands of plastic-wrap-like slices, the nearly cellular resolution map revealed the network of layers, fibers and microcircuits of the woman’s brain.

While variation exists among brains, across ages and individuals, they have largely the same distribution of brain structures and anatomy, said author Alan Evans of McGill University’s Montreal Neurological Institute. There are “subtle shape changes among individuals,” but all atlases start from one representative brain and go from there.

The team was chiefly limited by computing power and capacity. To map the human brain with 1 micron spatial resolution, which has been done for mouse brains, the atlas would take up 21,000 terabytes of data — essentially rendering it impossible to navigate. By comparison, BigBrain, with its 20 micron resolution, comprises about a terabyte of data. Prior MRI-based atlases had resolution of 1 millimeter.

Richard Leigh, a Johns Hopkins neurologist, said he’s looking forward to test-driving BigBrain for his research on stroke recovery. With the microscopic detail available, Leigh can see which particular groups of neurons are growing through stroke treatment rather than just a general fuzzy area.

Evans was in Seattle on Wednesday working with the Allen Institute for Brain Science. Created by Microsoft co-founder Paul Allen, who has committed $500 million since its start in 2003, the Allen Institute has assembled a less-detailed human brain atlas of its own.

BigBrain is part of the European Union’s Human Brain Project that brings together specialists in neuroscience, medicine and computing to decipher the mysteries of the brain.

President Obama announced in April an initiative to map the human brain, describing it as a way to discover cures for neurological disease and strengthen the economy.