AI Maps Brain's Waste-Clearing System, Revealing Dual-Speed Fluid Flow Critical to Alzheimer's

Mapping the Glymphatic System with AI - Neuroscience News

Summary: A multidisciplinary neuroengineering study has broken a major imaging barrier by mapping the exact flow velocity of the brain's waste-clearing infrastructure. The research utilizes physics-informed artificial intelligence to decode magnetic resonance imaging (MRI) data, revealing the hidden mechanics of the glymphatic system, the fluid network that washes away metabolic wastes like the amyloid-beta proteins linked to Alzheimer's disease. The AI models uncovered a dual-speed drainage blueprint, demonstrating that protective fluid moves 50 times faster across the brain's outer surfaces than it does when trickling through deep brain tissue. When a person goes into deep sleep, waterlike fluid circulates around the brain, washing away metabolic waste that is linked to diseases such as Alzheimer's. This process, known as the glymphatic system, was first described in 2012 by Maiken Nedergaard -- a pioneering neuroscientist and co-director of the University of Rochester Center for Translational Neuromedicine. But questions remain about the system's mechanics -- notably, how quickly the fluid circulates around the brain. Studying the circulation within a living brain is difficult to do without causing irreparable harm to a subject. "You can put a microscope on a small patch of the brain and watch what's happening there with a lot of detail, and we've worked with that type of data in the past, but it's only a tiny view of the overall process," says Professor Douglas Kelley from URochester's Department of Mechanical Engineering. "If you want to image whole brains, an MRI is a great approach because it gives you a three-dimensional view. But an MRI has serious limitations too, the biggest of which is that it does not capture the fluid flow velocity, at least not for flows this slow." Kelley and his colleagues from URochester, Brown University, and the University of Copenhagen turned to artificial intelligence for help. In a new study published in Science Advances, they outline how they used physics-informed artificial intelligence to determine fluid flow velocities from magnetic resonance imaging (MRI) data. Using videos of dye spreading across brain tissue over time, the neural networks the researchers built were able to deduce how fast the fluid flows and how permeable the brain tissue is. The results showed that there are two main ways that the glymphatic system washes away particles in the brain such as the amyloid beta proteins linked to Alzheimer's disease -- and one of these ways is much faster than the other. The fast flow of the glymphatic system's waterlike fluid moves at a few microns per second around the brain's open regions such as the surface between the skull and the brain, while the slower flow of the waterlike fluid trickles through the brain's deep tissue at a rate about 50 times slower. So far, the researchers have been working to get baseline measurements of fluid flow in the brains of animals such as mice to inform the AI tools. In the future, they hope to be able to compare the fluid flow in healthy and sick brains as well as young and old brains, with aspirations to eventually study circulation in humans. "We're working hard toward being able to measure the flow of waterlike fluids in and around human brains because then the clinical applications get a lot more important and exciting," says Kelley. "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. Or we could check when somebody has been concussed to see whether the fluid circulation in their brain is disrupted. This study gets us a step closer." Funding: The research is supported by the NIH National Center for Complementary and Integrative Health and the NIH BRAIN Initiative. Kelley's collaborators on the study include Brown University PhD student Juan Diego Toscano, URochester computational scientist Yisen Guo, Brown University PhD student Zhibo Wang, URochester PhD student Mohammad Vaezi, University of Copenhagen Associate Professor Yuki Mori, Brown University Professor George Karniadakis, and URochester Assistant Professor Kimberly Boster. Author: Luke Auburn Source: University of Rochester Contact: Luke Auburn - University of Rochester Image: The image is credited to Neuroscience News Original Research: Open access. "MR-AIV reveals in vivo brain-wide fluid flow with physics-informed AI" byJuan Diego Toscano, Yisen Guo, Zhibo Wang, Mohammad Vaezi, Yuki Mori, George Em Karniadakis, Kimberly A. S. Boster, and Douglas H. Kelley. Science Advances DOI:10.1126/sciadv.aeb0404 Abstract MR-AIV reveals in vivo brain-wide fluid flow with physics-informed AI The circulation of cerebrospinal and interstitial fluid plays a vital role in clearing metabolic waste from the brain, and its disruption has been linked to neurological disorders. However, directly measuring brain-wide fluid transport, especially in the deep brain, has remained elusive. Here, we introduce magnetic resonance artificial intelligence velocimetry (MR-AIV), a framework featuring a specialized physics-informed architecture and optimization method that reconstructs three-dimensional fluid velocity fields from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI). MR-AIV unveils brain-wide velocity maps while providing estimates of tissue permeability and pressure fields, quantities inaccessible to other methods. Applied to the brain, MR-AIV reveals a functional landscape of interstitial and perivascular flow, quantitatively distinguishing slow diffusion-driven transport [∼0.1 micrometers per second (μm/s)] from rapid advective flow (∼3 μm/s). This approach enables new investigations into brain clearance mechanisms and fluid dynamics in health and disease, with broad potential applications to other porous medium systems, from geophysics to tissue mechanics.

Artificial intelligence reveals hidden fluid flow patterns in sleeping brains

University of RochesterMay 27 2026 When a person goes into deep sleep, waterlike fluid circulates around the brain, washing away metabolic waste that is linked to diseases such as Alzheimer's. This process, known as the glymphatic system, was first described in 2012 by Maiken Nedergaard-a pioneering neuroscientist and co-director of the University of Rochester Center for Translational Neuromedicine. But questions remain about the system's mechanics-notably, how quickly the fluid circulates around the brain. Studying the circulation within a living brain is difficult to do without causing irreparable harm to a subject. You can put a microscope on a small patch of the brain and watch what's happening there with a lot of detail, and we've worked with that type of data in the past, but it's only a tiny view of the overall process. If you want to image whole brains, an MRI is a great approach because it gives you a three-dimensional view. But an MRI has serious limitations too, the biggest of which is that it does not capture the fluid flow velocity, at least not for flows this slow." Professor Douglas Kelley from URochester's Department of Mechanical Engineering Kelley and his colleagues from URochester, Brown University, and the University of Copenhagen turned to artificial intelligence for help. In a new study published in Science Advances, they outline how they used physics-informed artificial intelligence to determine fluid flow velocities from magnetic resonance imaging (MRI) data. Using videos of dye spreading across brain tissue over time, the neural networks the researchers built were able to deduce how fast the fluid flows and how permeable the brain tissue is. The results showed that there are two main ways that the glymphatic system washes away particles in the brain such as the amyloid beta proteins linked to Alzheimer's disease-and one of these ways is much faster than the other. The fast flow of the glymphatic system's waterlike fluid moves at a few microns per second around the brain's open regions such as the surface between the skull and the brain, while the slower flow of the waterlike fluid trickles through the brain's deep tissue at a rate about 50 times slower. So far, the researchers have been working to get baseline measurements of fluid flow in the brains of animals such as mice to inform the AI tools. In the future, they hope to be able to compare the fluid flow in healthy and sick brains as well as young and old brains, with aspirations to eventually study circulation in humans. "We're working hard toward being able to measure the flow of waterlike fluids in and around human brains because then the clinical applications get a lot more important and exciting," says Kelley. "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. Or we could check when somebody has been concussed to see whether the fluid circulation in their brain is disrupted. This study gets us a step closer." The research is supported by the NIH National Center for Complementary and Integrative Health and the NIH BRAIN Initiative. Kelley's collaborators on the study include Brown University PhD student Juan Diego Toscano, URochester computational scientist Yisen Guo, Brown University PhD student Zhibo Wang, URochester PhD student Mohammad Vaezi, University of Copenhagen Associate Professor Yuki Mori, Brown University Professor George Karniadakis, and URochester Assistant Professor Kimberly Boster. University of Rochester Journal reference: Toscano, J. D., et al. (2026). MR-AIV reveals in vivo brain-wide fluid flow with physics-informed AI. Science Advances. DOI: 10.1126/sciadv.aeb0404. https://www.science.org/doi/10.1126/sciadv.aeb0404