AI can predict Alzheimer’s disease seven years before onset with 70% accuracy: Study

A new AI system can predict Alzheimer's disease seven years before the first symptoms appear, possibly opening the door to earlier treatment, according to a recent study by the US National Institutes of Health.

Using patients' past medical records as input to train the machine learning system, the AI ​​model was 70% accurate in predicting Alzheimer's disease 70% of the time and 80% on the day before diagnosis, the study found. In fact, when researchers added basic demographic details such as year of birth, gender, ethnicity and race, predictive accuracy increased to 90%.

“Over the past few decades, electronic health records have become a rich … source of information for understanding and predicting complex diseases, particularly Alzheimer's disease,” the study noted. The researchers led earlier studies using health records to track Alzheimer's progression, as well as models that classify or predict dementia from clinical data.

“Neurodegenerative disorders are devastating, diverse and challenging to diagnose, and the burden is expected to grow with aging. Among them is Alzheimer's disease, the most common form of dementia after age 65, characterized by memory loss and other cognitive symptoms that are costly and difficult for patients and caregivers.” The researchers wrote.

To conduct the study, researchers at the University of California-San Francisco compiled clinical data for more than 250,000 patients from a vast database of medical records collected from 1980 to 2021 from millions of people.

AI models were trained on 70% of patient records, which included both Alzheimer's patients and controls: people not diagnosed with the disease. The remaining 30% of all patient records were “saved” for use in the evaluation portion of the study.

AI has been able to predict the onset of Alzheimer's disease with high accuracy.

“These findings may support the hypothesis that Alzheimer's disease may be related to general aging or frailty that may occur in non-neuronal systems or that may accompany Alzheimer's,” the researchers wrote. High-level prognostic groups for risk factors for Alzheimer's disease, either differentially or together.

In particular, some of the early predictors cited in research as contributing to Alzheimer's risk include high levels of cholesterol and other fats in the blood, heart attack, dizziness, cataracts, and the breakdown of cartilage between bone joints.

Perhaps one of the most surprising findings is osteoporosis as a female-specific predictor of Alzheimer's risk.

“In the University of California Data Discovery Forum, individuals with osteoporosis … showed a faster progression of Alzheimer's disease compared to non-concussed individuals,” the study noted. .. contrasted with female controls.

This level of predictive power could be a game-changer in the current fight against Alzheimer's. There is no medicine. For future Alzheimer's patients, having years to live could lead to new ways to slow or stop the disease before it causes irreversible damage.

The lead researcher did not respond to requests for comment. Decrypt.

Edited by Ryan Ozawa.