New AI-powered research led by a Carle Illinois College of Medicine student offers a promising new way to identify imaging markers for attention-deficit/hyperactivity disorder in adolescents, potentially leading to more objective diagnosis protocols. The team’s method uses deep-learning tools to help spot specific differences in the brain scans of young people with ADHD versus adolescents without the condition.
“ADHD is extremely difficult to diagnose and relies on subjective self-reported surveys,” said first-year CI MED student Justin Huynh, a research specialist in the Department of Neuroradiology at the University of California, San Francisco. “There is definitely an unmet need for more objective metrics for diagnosis. That’s the gap we are trying to fill.”
The research study by Huynh and his team is the first to leverage deep-learning tools to pinpoint markers in the brain that are common in the specialized MRI scans of young patients with ADHD. The team tapped into the multi-institutional Adolescent Brain Cognitive Development (ABCD) Study, which includes brain imaging, clinical surveys, and other data on over 11,000 adolescents from 21 research sites in the U.S. With the help of AI, the researchers identified nine white matter tracts (fibers) in which the MRI signature showed elevated levels of a certain marker that coincides with symptoms of ADHD.
“Many people with ADHD go undiagnosed, falling through the cracks and failing to receive treatment, which greatly impacts their quality of life. Our findings might provide a promising first step towards building an objective, imaging-AI-based approach to diagnosis,” Huynh said. He recently presented the team’s research findings at the prestigious Radiological Society of North America (RSNA) conference in Chicago, attended by scientists from across the globe.
The team’s next steps include further analysis of the nine white matter tracts that showed the abnormalities to gain a better understanding of the underlying mechanisms. Huynh says there’s also room to expand the research to include other similar conditions. “The methodology we present is general and can be applied to brain imaging to study other neurodevelopmental disorders. This is an unexplored area with much potential,” he said.
With a background in computer science, Huynh is one of a group of physician-innovators at CI MED who see AI as an integral tool in unlocking new insights that advance human health and improve patient outcomes. “This project is an important step in my journey to use AI to change the way health care is done, to make it more equitable, accessible, and capable,” Huynh said. “The AI revolution is only just beginning, and it will certainly transform every aspect of health care over the coming decades. I hope to play a significant part in this transformation.”
The full research team includes Pierre F. Nedelec, M.S., M.T.M., Samuel Lashof-Regas, Michael Romano, M.D., Ph.D., Leo P. Sugrue, M.D., Ph.D., and Andreas M. Rauschecker, M.D., Ph.D. from the Center for Intelligent Imaging, Department of Neuroradiology, at the University of California San Francisco Health.
Editor’s note: For more information on Justin Huynh’s presentation at RSNA in December 2023, click here.
Huynh’s work has been featured in several media reports. Links are available below:
Imaging Technology News Online
RNSA News Release
AuntMinnie.com resources for Radiologists