Study Finds AI Can Detect Alzheimer’s a Decade Before Symptoms Appear

San Francisco, CA (Law Firm Newswire) February 15, 2018 – Scientists have developed artificial intelligence (AI) that can be trained to identify the early signs of Alzheimer’s disease nearly 10 years before symptoms begin to emerge.

A research team from the University of Bari in Italy developed an algorithm that can detect structural changes linked to the disease in different areas of the brain. They trained an AI using 67 MRI scans from the Alzheimer’s Disease Neuroimaging Initiative database at the University of Southern California in Los Angeles. Of those, 38 were from Alzheimer’s patients while 29 belonged to healthy individuals.

“Machine learning has shown promise for Alzheimer’s diagnosis,” commented nationally known estate planning attorney Michael Gilfix of Gilfix & La Poll Associates. “Early detection is crucial when it comes to the disease. With an earlier diagnosis, people can seek treatment sooner, while having more time to prepare for the future by getting their legal and financial affairs in order.”

The researchers trained the AI to analyze images and correctly identify the difference between diseased brains and healthy ones. They then tested the algorithm’s accuracy by having it process 148 brain scans that included healthy people, patients with Alzheimer’s and individuals with mild cognitive impairment that was known to eventually develop into full-fledged Alzheimer’s.

The AI was able to diagnose Alzheimer’s with an 86 percent accuracy rating. It could also correctly detect mild cognitive impairment 84 percent of the time. According to the researchers, the results showed that the algorithm has the potential to be an effective tool for early Alzheimer’s diagnosis before the disease’s clinical symptoms appear.

The team said they plan to continue developing the AI’s accuracy so that it is reliable enough to be used as a noninvasive early detection system for patients. Currently, the most effective way to diagnose Alzheimer’s early is through a costly, invasive procedure that is not accessible to everyone. There are also plans to study the algorithm’s uses for other neurodegenerative conditions like Parkinson’s disease.