An illustrative portrait of a man interwoven with textShare on Pinterest
Scientists may have found a way to predict the onset of Alzheimer’s early by using AI. AI Jasmin Merdan/Getty Images
  • Past studies have identified some Alzheimer’s early risk factors including age, family history, and genetics.
  • Researchers from the University of California San Francisco have used AI to identify several early risk factors to predict a person’s Alzheimer’s disease up to seven years before symptoms occur.
  • Scientists identified early risk factors affecting both men and women, as well as a few that were sex-specific including erectile dysfunction and an enlarged prostate for men and osteoporosis for women.

According to researchers, around the globe, there are about 69 million people with prodromal Alzheimer’s disease — when a person has signs of mild cognitive decline — and another 315 million with preclinical Alzheimer’s disease where symptoms have yet to develop, but brain changes signal the potential for disease development.

With numbers like these, it’s no wonder why researchers are constantly looking for new ways to lower a person’s risk of developing this type of dementia.

Past studies have identified some Alzheimer’s early risk factors including age, family history, and genetics.

Now, researchers from the University of California San Francisco have used artificial intelligence (AI) to identify several early risk factors to predict a person’s Alzheimer’s disease up to seven years before symptoms occur.

While scientists identified some early risk factors affecting both men and women, they also found a few that were gender-specific including erectile dysfunction and an enlarged prostate for men and osteoporosis for women.

The findings were published in Nature Aging.

According to Alice S. Tang, an MD/Ph.D. student in the Sirota Lab, part of the Department of Pediatrics and the Bakar Computational Health Sciences Institute at the University of California San Francisco, and lead author of this study, spotting early risk factors for Alzheimer’s disease is important because as a neurodegenerative disease, by the time there is devastating symptoms and progression, it is a lot harder to “treat” in the sense of reversing that damage.

“Therefore, being able to identify at-risk individuals before decline is helpful not only to address possible modifiable risks but also for the development of [preventive] treatments and approaches,” Tang told Medical News Today.

“We would like to be careful with the wording here when we say ‘early risk factor’,” she continued.

“In general, we can use data to identify early associations​ that influence risk, but whether something is a causal risk factor or modifiable risk factor will require future investigation. Nevertheless, the first step, identifying those associations and coming up with hypotheses, is what our study contributes,” she added.

For this study, Tang and her team used AI to search a clinical database of more than 5 million people for co-occurring conditions in patients who had been diagnosed with Alzheimer’s disease.

“AI is a very broad term (that) often refers to the science of machines being able to think like humans and make decisions,” Tang explained. “Here we are applying a type of AI called machine learning where computers can learn from data. In general, we wanted to be able to account for large amounts of heterogeneous data — the clinical records — that may be highly correlated or may consist of many relationships.”

“We use AI to be able to account for this complexity, and we choose to pursue interpretability so that our model is not a ‘black box’ AI model, but one that can tell us what these early risk factors are in the AI decision-making so a clinician looking at the results can also choose to believe in the AI or not depending on those factors,” she added.

By using AI and the clinical database, the researchers identified several early risk factors for Alzheimer’s disease seen in both men and women, including high blood pressure, high cholesterol, and vitamin D deficiency.

From this information, the scientists found they could identify the people who would develop Alzheimer’s disease up to seven years prior with 72% accuracy.

“Since we focus here on future risk prediction as opposed to deciding whether someone has Alzheimer’s disease now, we anticipated our prediction problem (would) be harder, so we were not expecting perfect performance,” Tang said.

“The margin of error on the model can be viewed as a good thing because as an early prediction model that means there is (a) margin for us to change that risk. Nevertheless, we also expected some predictability due to the influence of already known risks — e.g., age and age-related conditions — which (are) not as modifiable,” she added.

In addition to the early risk factors identified in both men and women, researchers also discovered a few gender-specific risk factors.

These include erectile dysfunction and an enlarged prostate in men, and osteoporosis in women.

These findings add to other research concerning these three medical conditions and their possible link to an increased risk for Alzheimer’s disease or dementia.

Research published in June 2015 found an increased risk for both Alzheimer’s disease and non-Alzheimer dementia in people with erectile dysfunction.

A study published in February 2021 discovered that men with an enlarged prostate were persistently at a higher risk for developing Alzheimer’s disease and all-cause dementia.

Research published in December 2021 reported that osteoporosis may increase a person’s likelihood of developing Alzheimer’s disease or Parkinson’s disease in adults 40 years of age and older.

“One thing to emphasize is that these are risks we see at a population level, but for an individual, the model takes into account the combination of diseases,” Tang explained.

How to reduce Alzheimer’s risk

“Nevertheless, a doctor can take the population level risks to counsel patients to aim to control their cholesterol, engage in exercise, take plenty of calcium/vitamin D, or treat osteoporosis to minimize the influence of those diseases on the risks.”
— Alice S. Tang, lead author

“Ultimately, in the future, we imagine a personalized model in the clinic will be able to not only predict risk but also list the risk factors for the individual patient in front of the doctor for more targeted advice and treatments,” she added.

After reviewing this research, Dr. Karen D. Sullivan, a board certified neuropsychologist, owner of I CARE FOR YOUR BRAIN, and Reid Healthcare Transformation Fellow at FirstHealth of the Carolinas in Pinehurst, NC, told MNT she was pleased to see the massive amount of longitudinal data kept in electronic medical records being analyzed and put to good use with predictive models.

“What’s unique about this study is that it illustrates the gene-environment connection. One’s APOE genetic status becomes much more clinically meaningful in the context of two clear risk factors: high cholesterol and osteoporosis, especially in women,” Dr. Sullivan said.

“Hopefully, studies like this will reinforce to doctors and the public that patients do have some control about the aging of their brains. People want this information, they want to make the best choices every day that they can, but science-based brain health does not trickle down to the everyday person anywhere close to enough,” she added.

MNT also spoke with Dr. David Merrill, a geriatric psychiatrist and director of the Pacific Neuroscience Institute’s Pacific Brain Health Center in Santa Monica, CA, about this study.

Dr. Merrill commented that it is great to see helpful data coming out of the early use of AI as it’s applied toward identifying those at risk for Alzheimer’s disease sooner rather than later. He also touched on the importance of identifying early risk factors that can potentially be modified, such as high cholesterol and bone health.

“We know that in some ways our cholesterol is set by our genetics, which we don’t have control over, but we do have control over things like saturated fat content in our diets, which can affect cholesterol levels,” Dr. Merrill explained.

“We know that there are activities such as weight-bearing activities, strength training, improvements in diet and health, and also medical treatments that can address early bone loss or osteopenia to slow down and or prevent osteoporosis. So those efforts can also now be tied to greater brain health with aging, especially in women, which as we know the majority of persons living with Alzheimer’s disease are women.”
— Dr. David Merrill

“It’s particularly important that we identify modifiable risks in women who are vulnerable to developing Alzheimer’s with aging,” he added.