January 9, 2017 | Written by: James Kozloski, Ph.D.
Categorized: IBM Research
Share this post:
Understanding the brain is one of the grandest challenges we have in science. By understanding the brain, we will understand how the brain creates our cognitive processes, how these processes are implemented in brain tissues, and how the brain differs from other systems we see and build.
I’m a neuroscientist and have worked at modelling the brain for the past 15 years at IBM Research. I am named as an inventor on more than 80 patents.
Two of my patents that issued in 2016 were: Patent # 9,504,386, “Controlling Devices Based on Physiological Measurements” and Patent # 9,384,661, “Cognitive Needs-Based Trip Planning.” Both describe inventions that deliver cognitive computing technology to enhance a user’s day-to-day life by analyzing and predicting their next cognitive physiological state.
These two patents are among 8,088 patents awarded to IBM inventors in 2016, as the company positions itself for leadership in the new era of cognitive computing.
The work that my team of computational neuroscientists is doing in brain research is already contributing to understanding and enhancing how humans use technology, because our goal is to affect not just our field, but the future of our company. Today, we’re also aiming to make all our lives better by understanding brain disease.
Consider that every 67 seconds in the United States someone is diagnosed with Alzheimer’s Disease. It’s the sixth leading cause of death in the U.S.; and in fact, it’s the only killer disease in the top 10 that cannot be prevented or stopped.
I would argue we don’t even know what the disease’s basic biological mechanism is. Tens of billions of dollars have been spent in the public and private sectors on the science of brain diseases, and yet there remains a core problem: we don’t really understand what the brain is doing.
The way we’ve traditionally approached brain research and preventing or curing neurological diseases such as Alzheimer’s has led us in the wrong direction” instead. Instead of ignoring this core problem, what the brain is doing as a whole needs to inform our research and our questions of why the brain fails in specific places when its genes and brain cells called neurons go wrong.
With that insight, a new model has emerged. What I discovered is that the brain is not a collection of components arranged like a pipeline in a computer with information entering through our senses, kicking off transformations over a series of stages and then exiting through our muscles. Instead, it’s best modeled first as a closed loop.
Very little progress has been made in the past 20 years since the genetic cause of Huntington’s Disease was discovered. Like Alzheimer’s, Huntington’s is a devastating neurodegenerative disorder. But unlike Alzheimer’s, it’s caused by just one mutant gene.
I’ve proposed that instead of being toxic to brain structures, as is commonly believed, this gene’s mutant protein disturbs the brain’s internal dynamics in a subtle and risky way. Kind of like how using the wrong oil in your car might make it more likely for different parts to fail as you drive it over the years.
And to demonstrate this hypothesis, we needed to render my new model of brain circuitry in a computer simulation. So, 18 months ago, my team’s task became to study the circuit and discover its dynamics and rules — simulating the brain as it is and as it actually works on its own without inputs — then, identify the risks of neurodegenerative diseases and new strategies to eliminate them.
We call the core circuit of the model the brain’s “Grand Loop.”
Today my team is simulating this circuit and its processes and identifying those risks. We’re specifying the rules that govern brain dynamics, just as grammar governs the words I’m using now.
We’ve demonstrated the brain model for clients, for Huntington’s disease research foundations, and we published the model in early 2016 in the journal, “Frontiers in Neuroanatomy.” This may be the beginning of a new kind of neuroscience.
In addition to Alzheimer’s and Huntington’s, the brain model we’ve proposed can also address other diseases like Parkinson’s, and disorders like schizophrenia, PTSD, OCD, depression and chronic pain.
Cognitive healthcare is my company’s moonshot. Needless to say, it’s a fantastic time to be working as a neuroscientist on the brain’s mission.
How often does a single model, a single technology hold the promise to both transform information technology and save lives by providing insights into therapies for horrible diseases? This is what it means to be a scientist, reinventing neuroscience at IBM.
Also read, Why We Patent.
And to learn more about the new era of business, visit ibm.com/cognitive.