AI Maps the Brain's Waste-Clearing 'Fast Lane' and 'Slow Lane'

Using physics-informed AI trained on MRI data, researchers have for the first time measured how fast cerebrospinal fluid flows through a living brain — revealing a fast route near the surface and a deep-tissue route about 50 times slower. The technique could one day reveal circulation problems tied to Alzheimer's, aging and concussions.

Key Takeaways

Using physics-informed AI trained on MRI data, researchers have for the first time measured how fast cerebrospinal fluid flows through a living brain — revealing a fast route near the surface and a deep-tissue route about 50 times slower.
The technique could one day reveal circulation problems tied to Alzheimer's, aging and concussions.

AI synthesis2 sources8

During deep sleep, a water-like fluid flows around the brain, flushing away metabolic waste linked to diseases such as Alzheimer's. First described in 2012 by pioneering neuroscientist Maiken Nedergaard, this glymphatic system has remained frustratingly hard to study in living humans — above all, how fast the fluid actually moves.

Now, an international team led by the University of Rochester has answered that question with the help of physics-informed artificial intelligence. In a study published in Science Advances, researchers trained neural networks on videos of dye spreading through brain tissue, then used the AI to compute fluid flow speeds from MRI data — something conventional MRI cannot do for flows this slow.

The results reveal two very different routes. In open spaces near the surface of the brain, between the skull and the brain itself, the fluid travels at a few microns per second. Deep inside brain tissue, however, it moves about 50 times slower.

"The differences in flow speed were dramatic," said Douglas Kelley, professor of mechanical engineering at Rochester and senior author of the study, which included collaborators at Brown University and the University of Copenhagen. "If you want to image whole brains, an MRI is a great approach — but it does not capture the fluid flow velocity, at least not for flows this slow."

The team is currently using mice to establish baseline measurements of normal fluid flow, refining the AI tools as they go. The long-term goal is to extend the method to people: measuring fluid circulation in and around a human brain could allow clinicians to detect circulation problems associated with Alzheimer's, screen for poor circulation earlier in life, or check whether a concussion has disrupted the brain's cleaning system.

"We hope to someday be able to see whether an Alzheimer's patient has poor circulation in their brain or even screen for poor circulation earlier in life to try to stave off Alzheimer's," Kelley said. "This study gets us a step closer." The research was supported by the NIH National Center for Complementary and Integrative Health and the NIH BRAIN Initiative.