“BigBrain,” a 3-D brain model 50 times more detailed than anything that came before, is the result of an international collaboration led by 14 scientists from Germany and Canada, who announced completion of their work in a report published today in Science.
Katrin Amunts, Ph.D., the director of the Cecile and Oskar Vogt Institute for Brain Research at Heinrich Heine University Düsseldorf in Germany, led the project.
"With the help of our high-resolution brain model, we can now gain new understanding of the normal structure of different functional areas of the brain, such as the motor cortex, and regions that are important for learning and memory," Amunts explained in a press release.
It's nearly impossible for humans to grasp the immense interconnectivity of the brain, with its estimated 86 billion neurons and the same number of glial, or non-neuronal cells.
In fact, no array of super computers that exists today is capable of interactively exploring such a large amount of data. To put it into perspective, a brain scanned to a resolution of 1 micrometer (mm) would contain approximately 21,000 terabytes of data; one terabyte is equal to one trillion bytes!
For this reason, the researchers who created BigBrain were limited to a 20-mm resolution in three dimensions, still too large to see some cell structures, but far more refined than any previous digital model of the structure and function of the brain.
Thousands of Tissue Samples, One BigBrain
Amunts and her colleagues leveraged current advancements in computing and image analysis in a labor-intensive process that began in Düsseldorf five years ago.
Using a special tool called a microtome, the researchers carefully cut the wax-covered brain of a 65-year-old woman into more than 7,400 slices 20-mm thick, comparable to a single sheet of kitchen plastic wrap. Each section was stained with dye used to detect cell bodies and then digitally scanned.
The scanned brain sections were then painstakingly aligned and reconstructed. The result is a 3-D reference atlas 50 times more detailed than anything available in the past.
When asked at a press conference why they chose the brain of a 65-year-old, the researchers explained that this donated brain met all of their agreed-upon criteria—a healthy brain from a person with no neurological or psychiatric issues.
Although the essential brain structure varies little from person to person, the researchers plan to repeat the process on other brains to see just how much they differ from this one.
The Brain Is Today’s 'Big Science'
Advancements in neuroscience and computation are fueling brain research across the globe. In April, President Obama announced his intention to fund North American brain research on a larger scale, while Europe and China have raced ahead.
No previous brain map has probed deeper than the macroscopic, or visible, level. To achieve a unified understanding of the brain, from genes to cognition to behavior, a high-resolution model with microscopic components is required.
Recent innovations in neuroscience now allow scientists to integrate connectivity data into models of the brain's anatomy. BigBrain provides a scaffold-like structure for integrating this newly available molecular data into the reference brain.
It will allow scientists to assign molecular and gene expression information to features that were previously only visible under a microscope, paving the way for important insights into the biological basis of thought, language, emotions, and other brain processes.
Clinical Applications for BigBrain
Exploring the human brain is an international collaborative effort. Public access to the BigBrain dataset will be available for free through the CBRAIN Portal, the researchers said.
From a clinical standpoint, the opportunities a better reference model offers are limitless. Today, for example, patients with Parkinson’s disease are receiving deep brain stimulation implants to help control their tremors. BigBrain will allow for more precise placement of the electrodes at the brain sites affected by the disease.
Alzheimer’s research will also be aided by a more exact understanding of the target areas the disease affects in the brain. At a press conference, the researchers stressed that even today's best MRI images cannot identify the single nerve fibers or nerve bundles that are so important to understanding the processes of learning and memory.
The researchers also said they plan to take measurements of the brain's cortical thickness to gain insights into aging and neurodegenerative disorders. They will create cortical thickness maps to compare data from live brain imaging, and then integrate gene expression data from the Allen Institute for Brain Science in San Francisco, Calif.
Eventually, Amunts and colleagues hope to build a brain model at a resolution of 1 micrometer to capture details on the level of single cells. Meanwhile, the current version of BigBrain will help scientists better understand certain disease pathways and inform drug development.
Because of the sheer volume of data in BigBrain, the researchers say there will be a push by computer scientists who want to use it to develop new tools for visualization, data management, and analysis.
New computing structures and applications are already being developed by the European Human Brain Project.