Canadian Scientists Built a Virtual Map of the Human Brain

A video demo of the BigBrain project. Based largely in Montreal, scientists created a highly-detailed cross-sectional map of the human brain. Researchers can quickly zoom to any region they like for further analysis.

Your brain is a complex and marvelous creation. One hundred billion interconnected nerve cells, entangled in a bewildering network of 100 trillion synapses. These connections are where all of your memories are stored, from your first awkward sexual fumble in Grade 8 band camp to last week’s awkward sexual fumble with the weirdo you brought home from the after-hours you should have never been at in the first place.

That’s an incredible amount of neural data to be packed into 1.5 kilograms of mushy tissue. The storage security of all that information is surprisingly resilient—even when it’s challenged by massive amounts of liquor, hard drugs, or your many failed attempts to pull off an ollie 900.

But your mental data doesn’t always survive the abuse. Derek Boogaard, an enforcer with the New York Rangers, spent his whole career taking vicious shots to the head. In the weeks before his death, already heavily addicted to opioid painkillers, Boogard descended into wild fits of mania and depression. He overdosed on booze and pills and died just shy of his 29th birthday.

Boogaard’s family donated his brain to medical research. When it was examined, scientists found the tissue within was severely degraded—it was literally full of holes. He’d become a victim of chronic traumatic encephalopathy (CTE), a degenerative disease mainly affecting people who sustain repetitive brain injuries, a consistent profile for an athlete who got his head bashed in at regular intervals by Matt Carkner.

The symptoms of CTE are memory loss, confusion, impaired judgment, aggression, depression, and dementia. If this list of maladies sounds familiar, that’s because it’s shared with a whole host of other brain disorders, including Alzheimer’s Disease.

The structure of the brain, at a macroscopic level, has been known for a long time. But changes that affect our thinking patterns always begin at the cellular level. That’s why Canadian scientists have built a new map of our favourite organ from the ground up—almost cell by cell.

I spoke to Dr. Don Stuss, the President and Scientific Director of the Ontario Brain Institute, about how impressive and important this project really is: "This new model is a move from the macro to the micro level. This has enabled us to map—from the cellular level, from the basic core of what brain function is—to the macro level, like you'd get in a hospital scan… It's a phenomenal amount of work to achieve this"

The massive amount of brain power that's been put into developing the “BigBrain” project in the first place is an international collaboration, and as is often the story with multinational projects—like WWII—the Canucks got called in to do the dirty work. In this case, the intrepid folks at the Montreal Neurological Institute got their hands on a fresh brain, donated by a 65-year-old woman, with which they modeled their virtual brain map.

To start, they embedded the entire brain in a block of paraffin wax. Then came the hard part: using a microtome (a device much like a deli meat slicer, except equipped with a diamond-tipped blade), the entire block was sliced into segments 0.02 millimetres thick, like a gruesome smorgasbord of brain carpaccio. Each slice—some 7,400 in all—was numbered, fixed in a chemical brew to outline the cellular structures, and then digitized into a huge database via a flatbed scanner manned by hapless undergraduates.

7,000 of these thin human brain slices were used to create the BigBrain.

The highly-detailed digital images of each slice were added together to create a 3D map of the entire brain, at a resolution 50 times greater than what’s been previously achieved. The final result is BigBrain—a collaborative online tool. A researcher who's interested in looking at, say, the upper 1\10th of the amygdala now merely has to log into BigBrain and feed in some spatial coordinates. A few moments later, a hi-res map of just that section of grey matter pops up on screen. That means that a doctor in Marrakesh, who has discovered an abnormality in a particular section of a patient's brain, need only access the server in Montreal to see how the diseased variant differs from the digital display model. This is a very cool development in the collaborative world of science.

Having an instant brain reference model could lead to a renaissance in how we diagnose and treat neurological disorders. Once it's discovered that a particular disease is closely linked to a specific change in structure, in a particular region of the brain, a patient need only have an MRI scan done of their own noggin to see if they're carrying that particular defect. That would allow doctors to make a sharp diagnosis, quickly.

The BigBrain project is yet another example of the power of Big Data. Ten years ago, when computing power was expensive, scientists had to be very selective about what they were looking for. This approach was thorough, but it was easy to miss important details. Nowadays, a Samsung Galaxy is powerful enough to run nuclear weapons simulations, so researchers are scooping up all the data they can get their hands on. The challenge is to design smart software to separate the digital wheat from the virtual chaff.

Canadians are especially good at this sort of thing. Around 2000, when the Human Genome Project was nearing completion, the major challenge was storing and processing the immense amounts of data that were being generated. Fifteen years ago, when the thought of a one-terabyte hard drive was the stuff of a madman’s dreams (today’s list price: $100), researchers looking to comb through large amounts of information needed to design smart algorithms to look for meaningful patterns in the huge swaths of 1s and 0s their projects were generating. That kind of thinking was pioneered at places like The Centre for Applied Genomics at Sick Kids hospital in Toronto. Good thing too, because as computing power has skyrocketed in the last decade, those kinds of smart algorithms—which are getting more powerful and effective day by day—are finding new kinds of patterns.

An unexpected bonus of Big Data is that these algorithms are finding patterns by themselves that no one specifically intended to look for. As the amount of data that we generate goes stratospheric—consider a pair of Google Glasses constantly measuring your heart rate and temperature, for example—scientists will begin to see trends they never anticipated. For example, say in five years, when Rob Ford is still the duly elected mayor of Toronto, health authorities detect a big increase in body temperature among hundreds of revelers at Ford Fest, his honour’s annual BBQ. Suddenly, dozens of members of Ford Nation fall violently ill. Medical algorithms trace the confluence of those data points back to party and the mayor’s brother Doug is brought in on charges of spiking the steak sauce with the city’s finest crack cocaine. Hey, it could happen.

The idea here is that as more data floods into our servers, we will begin finding connections between things we once thought were totally unrelated—by sitting back and letting machines do the work for us.

We can only hope the next generation of Derek Boogaards who are suffering from severe brain trauama will benefit from the knowledge brought by BigBrain and Big Data, to ensure they stay healthy enough to let the blood blossom forth from their opponents when the gloves fly off.

Paul Tadich is a science journalist and a huge nerd who would love to take you to a dark tavern and talk to you in earnest, yet gentle tones about quantum vortices. You can visit his blog at paultadich.com

More science-y stuff:

Ray Kurzweil: That Singularity Guy

The Great Brain Data Pain

Sucking Up Radiation From Outer Space in the Chilean Desert