Researchers Use AI to Predict Alzheimer’s 7 Years Before Symptoms Appear

70 total views

Researchers at the University of California San Francisco say they have found a new way to use artificial intelligence to spot indicators of Alzheimer’s up to seven years before symptoms appear.

Researchers utilized a tool called Scalable Precision Medicine Oriented Knowledge Engine, or SPOKE, developed at UCSF “identify patterns and potential molecular targets for therapy.”

AI was used to look for “co-occurring conditions in patients,” utilizing more than 5 million electronic health records in the UCSF database. Of those, 2,996 individuals with Alzheimer’s had undergone evaluations at the Memory and Aging Center and “had expert-level clinical diagnoses.”

Researchers said they were able to identify who would develop the disease up to seven years prior with 72% predictive power.

High cholesterol for both men and women and osteoporosis in women were the most influential factors determining whether someone would later develop Alzhemer’s, according to the researchers..

“This is a first step towards using AI on routine clinical data, not only to identify risk as early as possible, but also to understand the biology behind it,” Alice Tang, an MD/PhD student in the Sirota Lab at UCSF and lead author of the study said in a University of California San Francisco online article. “The power of this AI approach comes from identifying risk based on combinations of diseases.”

The findings from the study were published in Nature Aging on Feb. 21.


Other factors that could determine Alzheimer’s risk for both men and women included hypertension, high cholesterol and vitamin D deficiency. Erectile dysfunction and enlarged prostates were predictive in men, and osteoporosis was predictive in women.

“It is the combination of diseases that allows our model to predict Alzheimer’s onset,” Tang said in the release. “Our finding that osteoporosis is one predictive factor for females highlights the biological interplay between bone health and dementia risk.”

The researchers said they hope to use the approach to identify other diseases, including lupus and endometriosis.

Share this Post