New AI platform accelerates brain inflammation research 10,000-fold
A new AI tool developed by Children’s National and Howard University analyzes brain immune cells 10,000x faster than manual methods.
A new Machine Learning and Artificial Intelligence tool from researchers at Children’s National Hospital (CNH) and Howard University (HU) accelerates discoveries in brain inflammation. Called StainAI, it rapidly and accurately analyzes microglia, the brain’s immune cells. Scientists currently analyze microglia slowly by hand. StainAI automates this process and speeds it up 10,000-fold. Its use will aid discovery of new treatments for inflammatory brain conditions such as infection, autoimmunity, and aging.
Solving a problem
Traditionally, scientists study microglia one cell at a time. They reconstruct each cell’s shape by hand under a microscope. The shape helps classify microglia as “resting” (normal) or “activated” (inflamed). The manual process is tedious and slow. It limits analyses to a few microglia in small brain areas.
StainAI changes that. It uses deep machine learning and artificial intelligence to overcome and exceed the manual method’s limitations. It correctly classifies millions of microglia from standard microscopic images. StainAI also localizes each microglia to its brain region in 3D. These features enable single-cell analyses of immune activity at a scale not feasible before – the entire brain.
A tool with broad impact
The team applied StainAI to two models of brain injury and inflammation to show its utility. In a rodent model of pediatric cardiac arrest, StainAI identified new brain regions susceptible to injury. In a simian model of viral infection, StainAI localized rod-shaped microglia normally found in white matter to an unexpected brain region – the hippocampus. These findings point towards new treatments and highlight StainAI’s value across diseases and species.
StainAI is fast, accurate and adaptable. It uses common laboratory equipment. Its creators, Michael Shoykhet, MD, PhD, at CNH and Dr. Tsang-Wei Tu at HU, are making StainAI available to other researchers. They hope StainAI will help labs worldwide discover new ways to protect children’s brains from inflammation and injury.
You can read the full article, StainAI: quantitative mapping of stained microglia and insights into brain-wide neuroinflammation and therapeutic effects in cardiac arrest, in Communications Biology, a Nature group journal.
