SAN FRANCISCO — Could a computer detect the first subtle signs of Alzheimer’s disease long before doctors diagnose the patient? Scientists have developed a cutting-edge technology that could transform how we identify Alzheimer’s years before traditional symptoms become obvious. Using nothing more than video footage of mice, researchers at Gladstone Institutes have created a machine learning tool (a form of AI) that can detect subtle behavioral changes that might signal the earliest whispers of brain dysfunction.
Imagine a world where a simple video recording could reveal the first, nearly invisible signs of a devastating neurological disease. That’s precisely the promise of VAME (Variational Animal Motion Embedding), a sophisticated computer algorithm that can spot behavioral irregularities invisible to the human eye.
“We’ve shown the potential of machine learning to revolutionize how we analyze behaviors indicative of early abnormalities in brain function,” says Dr. Jorge Palop, the study’s senior author, in a media release.
Unlike traditional medical tests that require complex tasks or expensive equipment, this technology can work with smartphone-quality video.
Published in the journal Cell Reports, the researchers studied two groups of genetically modified mice designed to simulate different aspects of Alzheimer’s disease. Instead of forcing the mice through predetermined tests, the team simply recorded their natural movements in an open arena. The machine learning tool then analyzed these recordings, revealing fascinating insights.
What did VAME discover?
As the mice aged, the system detected a significant increase in “disorganized behavior.” This doesn’t mean the mice were simply moving differently — they were showing more erratic patterns of activity, frequently switching between tasks in ways that might indicate emerging memory and attention problems.
“I envision this technology will be used to assess patients in the clinic and even in their homes,” explains Dr. Stephanie Miller, the study’s first author. “It gives scientists and doctors a way to solve the very hard problem of diagnosing preclinical stages of disease.”
The study didn’t stop at detection. The researchers also tested a potential treatment approach by blocking a specific blood-clotting protein called fibrin, which previous research suggested might contribute to brain inflammation. Remarkably, this intervention dramatically reduced the abnormal behavioral changes in the Alzheimer’s mice.
“It was highly encouraging to see that blocking fibrin’s inflammatory activity in the brain reduced virtually all of the spontaneous behavioral changes in Alzheimer’s mice,” notes Dr. Katerina Akassoglou, another researcher involved in the study.
While the research is preliminary and conducted on mice, it represents a potentially revolutionary approach to understanding and potentially treating Alzheimer’s. By catching the disease’s earliest signals, doctors might one day be able to intervene much earlier, potentially slowing or preventing cognitive decline.
The team’s ultimate goal is clear. As Dr. Miller puts it, her aim is “to make this tool and similar approaches more accessible to biologists and clinicians in order to shorten the time it takes to develop powerful new medicines.”
Source : https://studyfinds.org/ai-alzheimers-invisible-signs/
