Artificial Intelligence Helps Show How the Brain's Fluids Circulate

A novel artificial intelligence-based technology for detecting fluid flow around blood vessels in the brain could have significant implications for finding treatments for disorders like Alzheimer's disease. The perivascular gaps that surround cerebral blood arteries convey water-like fluids and aid in waste removal. Fluid flow changes have been associated with neurological disorders such as Alzheimer's, small artery disease, strokes, and traumatic brain injuries, but they are difficult to assess in vivo.

Innovative AI Velocimetry Techniques Unveil Mysteries of Brain Fluid Flow

The University of Rochester Associate Professor Douglas Kelley led a diverse team of mechanical engineers, neuroscientists, and computer scientists in developing unique AI velocimetry techniques to properly determine brain fluid flow. The findings were published in the Proceedings of the National Academy of Sciences (1^✔ ✔Trusted Source
Artificial intelligence velocimetry reveals in vivo flow rates, pressure gradients, and shear stresses in murine perivascular flows

Go to source). "In this study, we combined some measurements from inside the animal models with a novel AI technique that allowed us to effectively measure things that nobody's ever been able to measure before," says Kelley, a faculty member in Rochester's Department of Mechanical Engineering.

Role of AI in Complexities of Brain Fluid Flow

The findings are the result of years of research undertaken by study co-author Maiken Nedergaard, co-director of Rochester's Center for Translational Neuromedicine. Previously, the research was able to perform two-dimensional investigations on fluid flow in perivascular regions by infusing microscopic particles into the fluid and analyzing their position and velocity over time. However, more detailed measurements were required to fully comprehend the system's complexities—and investigating such a crucial, fluid system is difficult.

To address this issue, the team teamed with Brown University's George Karniadakis to use artificial intelligence. They used existing 2D data with physics-based neural networks to generate unprecedented high-resolution views of the system.

"This is a way to reveal pressures, forces, and the three-dimensional flow rate with much more accuracy than we can otherwise do," says Kelley. "The pressure is important because nobody knows for sure quite what pumping mechanism drives all these flows around the brain yet. This is a new field."

Reference:

1. Artificial intelligence velocimetry reveals in vivo flow rates, pressure gradients, and shear stresses in murine perivascular flows - (https://www.pnas.org/doi/10.1073/pnas.2217744120)

Source-Medindia