A new AI system can predict Alzheimer’s disease up to seven years before symptoms appear, paving the way for earlier treatments, according to research recently presented by the US National Institutes of Health.
Using patients’ medical records as input to train the machine learning system, the AI model was 70% accurate in predicting Alzheimer’s seven years in advance and 80% accurate the day before diagnosis, the study found. In fact, its predictive accuracy improved to 90% when the researchers added basic demographic details such as birth year, gender, ethnicity and race.
“Over the past few decades, electronic health records have become a rich source of data that can be leveraged to understand and predict complex diseases, particularly Alzheimer’s disease,” the study noted. The researchers leveraged previous studies that used health records to track the course of Alzheimer’s disease, as well as models that classified or predicted a dementia diagnosis from clinical data.
“Neurodegenerative disorders are devastating, heterogeneous, and difficult to diagnose, and their burden in aging populations is expected to continue to increase. Among these diseases, Alzheimer’s disease is the most common form of dementia after age 65, and its hallmark symptoms, memory loss and other cognitive symptoms, are costly and burdensome for patients and caregivers,” the researchers wrote.
To conduct the study, researchers at the University of California, San Francisco, compiled clinical data from more than 250,000 people from its vast database of millions of people’s medical records collected between 1980 and 2021. Nearly 3,000 of those patients had been diagnosed with Alzheimer’s disease.
The AI models were trained on 70% of the patient records, which included both patients with Alzheimer’s disease and controls, meaning people who had not been diagnosed with the disease. The remaining 30% of the total patient records were “saved” for use in the evaluative part of the study.
AI was able to predict the onset of Alzheimer’s disease with a high degree of accuracy.
“These findings potentially support hypotheses suggesting that Alzheimer’s disease may be associated with general aging or frailty, which could manifest in non-neurological body systems before or concurrently with Alzheimer’s disease,” the researchers wrote. “Furthermore, interpretation of these models allows for the identification of clusters of higher-order predictors that may contribute to disease heterogeneity or, together, contribute to Alzheimer’s disease risk.”
Specifically, some of the earliest predictive factors cited in the research that contribute to Alzheimer’s disease risk were high levels of cholesterol and other fats in the blood, congestive heart failure, dizziness, cataracts and deterioration of the cartilage between bone joints.
Perhaps one of the most surprising findings was the identification of osteoporosis as a female-specific predictor of Alzheimer’s disease risk.
“In the University of California Data Discovery Platform, individuals exposed to osteoporosis… showed faster progression to Alzheimer’s disease compared to matched unexposed individuals,” the study noted, “When stratified by sex, this progression was significant when comparing women with osteoporosis… to control women.”
This level of predictive power could be a game-changer in the fight against Alzheimer’s disease, which currently has no cureHaving several years of advance warning for potential Alzheimer’s patients could lead to new ways to slow or stop the disease before it causes irreversible damage.
The lead researcher did not respond to a request for comment from Decrypt.
Edited by Ryan Ozawa.
