Alzheimer’s disease is a serious problem. It affects millions of people, especially older adults. It harms the brain and makes it hard for people to remember things and live on their own.
Right now, doctors can only diagnose Alzheimer’s after people start showing signs. The problem is that by that point, the disease has already hurt the brain a lot.
Using Health Records: An Unseen Treasure Trove
The University of Florida researchers thought of a different approach. They knew that electronic health records could hold a lot of useful information.
These records have details about people’s health, like if they are overweight, have high blood pressure, or have high cholesterol. These conditions are linked to Alzheimer’s.
The Power of Artificial Intelligence: Predicting Alzheimer’s
The team built an artificial intelligence (AI) system. They trained it to pick out this information from health records.
The goal was to find out if the AI could tell which people would get Alzheimer’s, even before they start showing signs. They tested their AI system and shared their findings in the Alzheimer’s & Dementia journal.
The results were very promising. Both AI systems they tested did a great job. They scored very high in tests that measure how well an AI can do a task.
Further Steps: Testing and Refining the System
However, more testing is needed before doctors and patients can start using these AI tools.
But, this study shows that it’s possible to use health records to screen for Alzheimer’s.
Digging Deeper: Understanding the Findings
The team used health records from about 17 million people in Florida. They focused on roughly 24,000 people over 40 who had been diagnosed with Alzheimer’s or dementia.
They also looked at around a million people over 40 who didn’t have dementia.
The researchers made two AI models and tested them. They checked how well these models could pick out the people who would later get Alzheimer’s.
They looked at how accurate the models were at the time of diagnosis and one, three, and five years before diagnosis.
One model based its predictions on what scientists already know about Alzheimer’s. It used information like health conditions, habits, and medications that could lead to the disease.
The other model was more flexible. It could consider other data from the health records that might be linked to Alzheimer’s.
Breaking New Ground: Unexpected Discoveries
The flexible model did a better job than the other model. It even found some potential risk factors that the other model didn’t. These included muscle weakness, mood disorders, and tiredness.
Interestingly, it also found that women who have regular health check-ups, including exams for women’s health, are less likely to get Alzheimer’s than women who don’t have these check-ups.
Looking Ahead: The Potential for Early Intervention
Screening tests like these could be a key first step to better treatment and outcomes for people with Alzheimer’s. Catching the disease early could mean that treatments might work better and even delay or prevent Alzheimer’s.
The researchers are hopeful. They believe that these AI models could predict Alzheimer’s disease up to five years before diagnosis.
This could provide a critical window for intervention and potentially change the future for those at risk of Alzheimer’s.
If you care about dementia, please read studies about how the Mediterranean diet could protect your brain health, and common nutrient in meat, fish and beans may be key to preventing Alzheimer’s.
For more information about brain health, please see recent studies that cranberries could help boost memory, and many older people have this non-Alzheimer’s dementia.
The study was published in Alzheimer’s & Dementia.
Follow us on Twitter for more articles about this topic.
Copyright © 2023 Knowridge Science Report. All rights reserved.
