Novel AI Now Able To Predict Alzheimer's Disease 7 Years Before Onset of Symptoms

While Alzheimer's disease still does not have a cure, early detection could tremendously help, as it implies being able to prepare and have preventive measures in place.

New AI Can Predict Alzheimer's

Now, new AI models could soon offer individuals early warnings regarding the development of symptoms years prior to their onset.

Researchers from the University of California, San Francisco as well as Stanford University made use of machine learning approaches to over five million health records. They trained the model to identify patterns that link Alzheimer's to other kinds of conditions. The development was noted in the "Leveraging electronic health records and knowledge networks for Alzheimer's disease prediction and sex-specific biological insights" study.

The system that resulted is not a perfect one. However, when it was tested against records of individuals who were known to have Alzheimer's later on, it was able to accurately predict Alzheimer's development in 72% of cases. In some cases, it was able to do so up to seven years before.

The predictive capacity of the model comes from its ability to mix analyses of several various risk types for calculations of the likelihood of Alzheimer's development. Findings could shed more light on the causes of the diseases and on the susceptibility or risk that certain individuals may have.

Bioengineer Alice Tang, who is from the UCSF, explains that this is a first step towards the use of AI in routine clinical data. It will help to not just distinguish risk as early as possible but also shed light on the condition's underlying biology.

The AI model was able to pick up different conditions that can be used for Alzheimer's risk calculation. These include vitamin D deficiency, high cholesterol, high blood pressure, and depression. An enlarged prostate and erectile dysfunction were also crucial factors for men, while osteoporosis was an important factor for women.

This does not mean that individuals who have such health issues will end up developing dementia. However, the AI notes that these are predictors that could be worth a look.

The researchers also looked into the biology that underlies some of the links that they were able to identify. Alzheimer's in women, osteoporosis, and the MS4A6A variant gene were found to be linked to each other. This offers new doors for studying the development of the disorder.

ALSO READ: Alzheimer's Mysterious Origins May Be Driven by "Exhausted" Immune Cells, Study Reveals

Alzheimer's Disease

Alzheimer's disease, which is the most common dementia type, is a progressive condition that typically starts with mild memory loss and possibly leads to the loss of the ability to respond to the environment and hold a conversation. The conditions involve brain regions that manage language, memory, and thought.

In 2020, around 5.8 million Americans reportedly lived with the condition. Such numbers are thought to reach nearly triple, equivalent to 14 million individuals, by the year 2060. Disease symptoms can appear after 60 years of age, with the risk going up with age.

While the exact causes of Alzheimer's still remain a mystery to the scientific community, age is considered the best known risk factor for the condition. Family history could also play a role. Researchers are also looking into whether environment, diet, and education could play roles in Alzheimer's disease development.