NEW YORK — A new AI tool developed by NYU researchers can effectively spot brain changes that happen from repeated head injuries. Specifically, it can pick up on injuries that traditional medical imaging misses. The team says that the technology can open doors for better understanding how subtle brain injuries impact cognition over time.
For a long time now, experts have known about the potential risks of concussions among athletes that play contact sports like football. Evidence is now starting to show that that repeated head impacts, even if they seem to be mild at first, may lead to cognitive decline over time. Advanced MRI can identify microscopic changes in brain structure from head trauma, but the scans yield a lot of data that can be hard to examine.
This new technology utilizes machine learning to accurately compare the brains of male athletes who played contact sports like football and non-contact sports like track and field. The researchers studied hundreds of brain images from 81 male college athletes, 36 contact sport players (mainly football players) and 45 non-contact sport athletes (mainly runners and baseball players). The team found repeated head impacts with small structural changes in the brains of contact-sport athletes. Most interestingly, none of the participants had been diagnosed with a concussion.
“Our findings uncover meaningful differences between the brains of athletes who play contact sports compared to those who compete in noncontact sports,” says study senior author and neuroradiologist Yvonne Lui, MD, in a university release.
“Since we expect these groups to have similar brain structure, these results suggest that there may be a risk in choosing one sport over another,” adds Lui, a professor and vice chair for research in the Department of Radiology at NYU Langone Health.
To bring the program to life, the researchers designed statistical techniques that allowed their program to “learn” how to predict exposure to repeated head impacts using mathematical models based on data examples given to them. As scientists inputted more data, the program became “smarter.” The program was also able to identify unusual features in brain tissue and tell the difference between athletes with and without repeated exposure to head injuries based on the features.
Mean diffusivity and mean kurtosis were the two metrics that most accurately flagged brain structural differences. The first metric, mean diffusivity, measures how easily water can move through brain tissue and is often used to identify strokes on MRI’s. Mean kurtosis examines the complexity of brain tissue structures and can identify changes in the parts of the brain responsible for learning, memory, and emotions.
This new program has the potential to literally change the game of sports medicine, unlocking new ways to prevent long-term cognitive issues in athletes of all sports.
“Our results highlight the power of artificial intelligence to help us see things that we could not see before, particularly ‘invisible injuries’ that do not show up on conventional MRI scans,” says study lead author Junbo Chen, MS, a doctoral candidate at NYU Tandon School of Engineering. “This method may provide an important diagnostic tool not only for concussion, but also for detecting the damage that stems from subtler and more frequent head impacts.”
Since this study only included male participants, Chen says the next step involves exploring the use of their machine-learning program on female athletes.
The findings are published in The Neuroradiology Journal.
You might also be interested in:
- ‘Super-resolution’ brain scans take imaging to next level, could help doctors detect disease sooner
- Youth tackle football connected to permanent brain damage, early decline
- A third concussion could be the final strike for healthy brain functioning