Machine Learning Sheds Light on Dementia Risk Factors

Machine learning, a hot field in computer programming, uses artificial intelligence to “teach” a computer how to analyze data by itself — without additional human programming. Machine learning is currently being used in every field, from cybersecurity to self-driving cars, so it should not be a shock it is also being applied to detecting patterns within a variety of diseases.

The first study to use machine learning to analyze dementia risk was published in the Journal of Alzheimer’s Disease. The data it used is one of the oldest and most respected in the country: the Framingham Heart Study (FHS). The FHS began in 1948, with a group of just over 5000 men and women from Framingham, Massachusetts. Under the direction of the National Heart, Lung and Blood Institute, it has followed three generations of Framingham residents and their spouses, revealing a goldmine of genetic and environmental information about cardiovascular disease.

However, this huge pool of data is useful for other diseases as well. The Boston University School of Medicine turned to FHS in order to find risk factors for dementia, particularly risk factors that can be identified before dementia shows its symptoms.

Moreover, the study sought factors that could be identified even by a family member, rather than relying on today’s standards of dementia screening, which require specialized training. If only a neurologist with advanced training can test for a disease makes it more likely that the disease has already progressed significantly. “The frontline for screening are primary care physicians or family members,” explains Prof. Rhoda Au, one of the study’s authors. “We wanted to identify information that any physician or even non-physician has easy access to in determining potential increased future risk for dementia.”

What did the study find? Older age was obviously a significant risk factor, but other, less obvious associations were also found. A lower BMI, normally thought to reflect better health, was associated with dementia, as was a marital status of “widowed,” and, in an intriguing finding: having had less sleep between the ages of 40 and 60.

Although the mechanisms behind these risk factors are not yet understood, knowledge of their existence can already direct clinical decisions. If a widow or widower is underweight or has a history of sleeplessness, they can be watched for early signs of cognitive change. These simple, readily observed factors can alert family and doctors to subtle cognitive changes that will manifest as full-blown dementia only later. And catching dementia early makes all the difference, since at its earliest stages it can most easily be managed.

At Hamilton Grove Healthcare and Rehabilitation Center, in Hamilton, NJ, we specialize in caring for individuals suffering from Alzheimer’s, dementia, or other cognitive disorders. In our safe and secure unit, our specially-trained caregivers provide extra sensitivity and understanding to offer programming that helps residents maximize their cognitive function. Our activities program also foster greater socialization and appreciation of life. In addition, we use state-of-the-art therapies, such as audiovisual therapy and aromatherapy to grant our cognitively-impaired residents a calming atmosphere.

Read our reviews on,, and to hear what our residents and their families have to say.

Or better yet, come see for yourself: Contact us to schedule a tour by calling 609-588-5800 or by clicking here.

Leave a Comment