Alterations in the complex fluid flow system of the brain are associated with neurological conditions such as Alzheimer’s disease, small vessel disease, stroke and traumatic brain injury. (Getty Images)
New research targets diseases including Alzheimer’s disease.
A new artificial intelligence-based technique that measures fluid flow around blood vessels in the brain could have a major impact on the development of treatments for diseases such as Alzheimer’s.
The perivascular space, which surrounds the blood vessels of the brain, transports water-like fluids around the brain to help remove waste products. Changes in fluid flow are associated with neurological conditions such as Alzheimer’s disease, small vessel disease, stroke and traumatic brain injury, but are difficult to measure in vivo.
A multidisciplinary team of mechanical engineers, neuroscientists, and computer scientists led by Associate Professor Douglas Kelly at the University of Rochester has developed a new AI velocimetry to accurately calculate brain fluid flow. The results are outlined in a study published by. Proceedings of the National Academy of Sciences.
“In this study, we combined some measurements from inside an animal model with new AI techniques, which allowed us to effectively measure things that no one had been able to measure before,” says Rochester. says Kelly, a university mechanical engineering faculty member.
The video shows the perivascular space (the area within the white line) where the researchers injected the microparticles. Particles (shown as moving dots) are followed by lines indicating their direction. After measuring the position and velocity of the particles over time, the team integrated this 2D video with a physics-based neural network to create an unprecedented high-resolution 3D view of the brain’s fluid flow system. created. (Douglas Kelly)
The study builds on years of experimentation led by study co-author Miken Nedergaard, co-director of the Center for Translational Neuromedicine in Rochester. The group was previously able to conduct two-dimensional studies of fluid flow in the perivascular space by injecting microparticles into the fluid and measuring their position and velocity over time. However, scientists need more complex measurements to fully understand the complexity of the system, making it difficult to explore such critical flow systems.
To meet this challenge, the team worked with George Karniadakis of Brown University to leverage artificial intelligence. They integrated his existing 2D data with a physics-based neural network to create an unprecedented high-resolution system look.
“It’s a much more accurate way of revealing pressure, force, and three-dimensional flow than other methods,” says Kelly. “Pressure is important because no one yet knows exactly what pumping mechanism drives all these flows in the brain. It’s a new field.”
The scientists conducted the research with support from the Computational Neuroscience Collaborative Research Program, the National Institutes of Health’s Brain Initiative, and the Army Research Service’s Interdisciplinary University Research Initiative Program.
read more
Category: feature
