Rensselaer Polytechnic Institute researchers are hoping to improve surgeons’ skills by tapping directly into their brains.
A $2.2 million dollar grant from the U.S. Army Medical Research and Development Command of the U.S. Department of Defense will help researchers at RPI work with colleagues at the University at Buffalo. They’re participating in a two-stage project combining neuroimaging, artificial intelligence, and neuromodulation to better understand and measure how surgeons acquire and hone their skills — and then determine if that mastery can be accelerated.
Suvranu De is the director of CeMSIM, Rensselaer's Center for Modeling, Simulation, and Imaging in Medicine.
"So basically, to understand this, you know clearly, so when you're learning a task, like, you know, learning to play a guitar, or a violin, or a sport, or surgery, right, we do things repetitively over and over again. And then we learn that, and that learning essentially is embedded in the brain, as you know, neural connections, which is known as neuroplasticity. So far, when we've been evaluating surgeons, or evaluating a pianist, or evaluating a sports person, essentially, what we've done is look at external measures, look at how the hand is moving, or how, you know, certain metrics of motion are being recorded. And that has worked well, except that we did not really know how many of those things we have to record to actually figure out whether somebody is becoming better at something."
Surgeons would complete technical tasks while their brain activities are monitored, or neuroimaged, to reveal how well they have mastered critical skills. De says researchers will analyze that data using a collection of deep learning algorithms — known as a deep neural network — to assess and quantify each individual’s level of learning and skill.
"So we decided to go directly and look at the human brain. And think of this, as, you know, pulse oximeter, which is pretty well known nowadays. They're measuring oxygen level, in the blood grade. So if we can actually figure out which regions of the brain are lighting up when it's being active, right, during the learning process, we could potentially start to understand how somebody learns, and how the learning progresses over time. And whether the distinctions between somebody who has learned a task versus somebody who is still novice at that."
In the project's second stage researchers will examine whether neuromodulation can affect neural activity to facilitate learning. Researchers will observe the brain activity of medical students performing technical tasks while neural stimulation is applied. De says subjects will wear special headgear that allows laser light to assist researchers in seeing what is happening in the brain.
"They will actually go through the skull, and the tissue directly into the brain, and then those are scattered by the turbidity of the brain, then we can actually pick up that scattered light, right, that comes reflected back from the brain tissue. And that's what gives us the indication as to what is happening in the brain, which parts of the brain are active, which are not, how the activity is changing, and so on."
Dr. Steven Schwaitzberg chairs University at Buffalo's Department of Surgery. He explains the research targets brain proficiency, that mindset you reach when you’re able to perform a task automatically, without thinking. His best example: how you felt when you first learned to drive a car, versus being a seasoned licensed driver.
"Let's say, you're a surgeon soldier, and you went to be deployed in some other country, and you come back a year later, we can actually measure possibly, possibly as the key word. Are you ready to go back and do this elegant elective surgery that you were doing before you left, document it, that you're good to go, or provide opportunities for training? If the skills have gotten rusty. Nobody has ever been able to do that before. We've got a long ways to go. This is not tomorrow's, you know, routine medical care. But this is the path that we're on."
De says that research is being conducted on brain computer interaction. So the foundation is being set to allow future researchers to develop the technology to record things from one brain and then create interactions with another brain.
"Our possibilities are endless, actually here, because now for the first time we're able to peer into the human brain, and figure out what it means to learn a task."
De calls the RPI/UBuffalo initiative a one-of-a-kind project that “stands at the bleeding edge of research at the interface of neuroimaging, deep learning, and surgical skill assessment.”