Researchers Utilize AI to Measure Fluid Flow in the Brain’s Glymphatic System

In a groundbreaking study, researchers have employed artificial intelligence (AI) to better understand how the glymphatic system, responsible for clearing metabolic waste from the brain, operates during deep sleep. This system, first introduced in 2012 by neuroscientist Maiken Nedergaard of the University of Rochester’s Center for Translational Neuromedicine, plays a critical role in removing waste materials linked to diseases like Alzheimer’s.

Despite its importance, the exact mechanics of the glymphatic system, particularly the speed at which cerebrospinal fluid flows through the brain, remain unclear and challenging to measure accurately without causing damage to brain tissue. Professor Douglas Kelley from the University of Rochester’s Department of Mechanical Engineering emphasizes the limitations of traditional imaging methods like MRI, which provide a three-dimensional perspective but fail to capture the slow fluid flow velocity adequately.

To address this limitation, Kelley and his colleagues from the University of Rochester, Brown University, and the University of Copenhagen developed a physics-informed AI method to analyze MRI data. Their recent publication in Science Advances details how they trained neural networks using videos of dye diffusion in brain tissue. This innovative approach allowed the AI to estimate fluid speeds and the permeability of surrounding tissues.

Two Distinct Flow Rates Identified

The study identified two primary pathways through which the glymphatic system eliminates particles, including amyloid beta proteins associated with Alzheimer’s disease, revealing that fluid movement occurs at notably different speeds. In areas of the brain closer to the surface, the fluid flows at several microns per second. However, deeper within the brain tissue, its movement is approximately 50 times slower.

Currently, the research team is using animal models, including mice, to gather baseline measurements of normal fluid movement. These benchmarks are essential for refining their AI tools. Future work aims to contrast circulation patterns between healthy individuals and those with neurological issues, as well as across various age groups.

Future Applications in Human Health

A significant aim of this research is the eventual application of these techniques in humans, potentially offering new methods to examine neurological diseases and brain injuries. Professor Kelley confirms that the capacity to measure fluid dynamics in human brains could hold substantial clinical significance, such as assessing circulation in Alzheimer’s patients or evaluating concussion impacts on fluid movement.

This study moves the field closer to understanding how altered fluid circulation might influence brain health and disease. The research has been supported by the National Institutes of Health’s National Center for Complementary and Integrative Health and the BRAIN Initiative. Collaborators on the project include Brown University PhD students Juan Diego Toscano and Zhibo Wang, University of Rochester students Yisen Guo and Mohammad Vaezi, University of Copenhagen’s Associate Professor Yuki Mori, Brown University’s Professor George Karniadakis, and URochester’s Assistant Professor Kimberly Boster.