AI automates process of identifying Alzheimer’s disease pathologies
An artificial intelligence tool, developed by a team of University of California researchers, is able to detect Alzheimer’s disease markers in autopsied human brain tissue.
UC Davis and UC San Francisco scientists used a database of tens of thousands of labeled example images to train a convolutional neural network (CNN) algorithm to quickly determine if brain tissue samples had amyloid plaque—clumps of protein fragments between the brain’s nerve cells that are one of the hallmarks of Alzheimer's disease.
In a proof-of-concept study published on Wednesday in the journal Nature Communications, researchers report that that their CNN algorithm could process an entire whole-brain slice slide with 98.7 percent accuracy and conclude that AI can augment the expertise and analysis of an expert neuropathologist.
“This scalable means to augment a neuropathologist’s ability suggests a route to neuropathologic deep phenotyping,” state the study’s authors. “Significantly, models trained using a comparatively small investment of a neuropathologist’s time can assist with new cases and potentially reduce overall expert burden.”
“We still need the pathologist,” says lead author Brittany Dugger, an assistant professor in the UC Davis Department of Pathology and Laboratory Medicine. “This is a tool, like a keyboard is for writing. As keyboards have aided in writing workflows, digital pathology paired with machine learning can aid with neuropathology workflows.”
Researchers have made the CNN model code and dataset publicly available to promote use of the AI tool.
“It’s notoriously hard to know what a machine-learning algorithm is actually doing under the hood, but we can open the black box and ask it to show us why it made its predictions,” says Michael Keiser, an assistant professor in UCSF’s Institute for Neurodegenerative Diseases and Department of Pharmaceutical Chemistry.
“It’s a co-pilot, a force multiplier that extends the scope of what we can accomplish and lets us ask questions we never would have attempted manually,” added Keiser. “For example, we can look for rare plaques in unexpected places that could give us important clues about the course of the disease.”