EMC helps Neuro to create 3D brain database

MONTREAL – The Montreal Neurological Institute announced that it has received $1 million worth of supercomputing data-storage technology from EMC Corporation. The donation will allow researchers to map the human brain to a degree of precision that would have been unthinkable a generation ago.

Alan Evans (pictured), director of the McGill Centre for Integrative Neuroscience, told the Montreal Gazette that the technology will usher in a new golden age in brain research at the MNI. “Montreal has been a capital of brain research from the days of (Wilder) Penfield,” Evans said, referring to the legendary neuroscientist who founded the MNI in 1934.

“What is happening right now is we are rebuilding our position at the top of the global tree in brain research using information technology and using what we call brain mapping – where you are no longer describing disease, you are quantifying it, you are measuring it, you are looking across hundreds or even thousands of individual brains to understand variability in the brain.”

At present, brain surgeons and neuroscientists often rely on the inky images of two-dimensional MRI scans. What Evans and his colleagues are seeking to accomplish is a three-dimensional atlas of the brain right down to the level of neurons.

That poses an enormous challenge, given that the brain contains about 100 billion neurons and 1,000 trillion neuronal connections. In order to achieve that goal, the MNI needs lots of supercomputing power and plenty of data storage.

That’s where EMC Corporation comes in, with its “Isilon scale-out NAS” technology. In a world first three years ago, Evans’ team and German neuroscientists produced the Big Brain, a 3D computerized atlas of the mind. The atlas was assembled from 7,400 actual slices of a brain embedded in paraffin wax. Each slice was as thick as a human hair.

For the next step, they plan to produce a 3D atlas that will be assembled from 148,000 slices of a brain. With such detail, one will be able to glimpse the neurons of every part of the brain in the three dimensions.

Without super-fast computing and data storage in the petabytes (each petabyte contains 1,024 terabytes or a million gigabytes), the project would be impossible.

To put the numbers in more context, the Big Brain atlas currently provides 125,000 times more resolution than an MRI image. And the next-generation atlas will provide 20 times more resolution than the Big Brain.

“When you start talking about healthcare and what Alan’s team is trying to accomplish, resolution is key,” said Mike Sharun, an EMC Corporation executive.

“So the better the resolution, the more data that we have and the better the results, and the better the healthcare we’re going to have.”

As for the practical applications, Evans said that neurosurgeons will be able to use their brain atlas in preparing for surgery and researchers will rely on it to gain a deeper understanding into diseases like Alzheimer’s and Parkinson’s.

“To give you a practical example, when people are doing deep brain stimulation for Parkinson’s, they have to put a probe deep into very small structures in the brain,” Evans explained. “With an MRI, you can barely see those structures. With an atlas like this, the Big Brain, it allows our surgeons to interpret the MRI at a much finer level of detail.

“Another example would be if you’re trying to understand Alzheimer’s disease. Well, Alzheimer’s disease has many elements to it, blood flow and the passage of toxic proteins through the brain. We have to model that and understand it in a way that can only be done if you have large-scale computational tools.”