A team of US researchers has developed a new artificial intelligence (AI)-based tool that can recognize subtle signs of Alzheimer’s disease that emerge decades before a formal diagnosis is made. Signs often take the form of erratic behavior that reflects the early stages of brain dysfunction.
The team at the Gladstone Institute in California engineered mice to mimic key aspects of Alzheimer’s and used new video-based machine learning tools to detect early signs of the brain disease.
The findings, published in the journal Cell Reports, highlight a new strategy for identifying neurological disease earlier than is currently possible and tracking how it develops over time.
Gladstone investigator George Palop said AI could potentially revolutionize how behaviors associated with Alzheimer’s are analyzed – indicative of early abnormalities in brain function.
A machine learning platform called VAME (Variational Animal Motion Embedding) analyzed video footage of rats exploring an open area. It identified subtle behavioral patterns – disorganized behavior, unusual patterns and more frequent transitions between different activities – as the mice aged. These behaviors, possibly related to memory and attention deficits, were captured on camera, but could not have been noticed simply by watching the rats.
The tool could help understand the origins and progression of devastating brain disorders, Palop said, noting that it could also be applied to other neurological diseases.
Additionally, the new study also used VAME to learn whether a potential therapeutic intervention for Alzheimer’s would prevent disorganized behavior in mice.
They found that genetically blocking a blood-clot-forming protein called fibrin that triggers toxic inflammation in the brain could prevent the development of abnormal behavior in Alzheimer’s mice.
The team said the intervention also tackled spontaneous behavioral changes in Alzheimer’s mice.
(Except for the headline, this story has not been edited by NDTV staff and is published from a syndicated feed.)
