Researchers have developed an artificial intelligence program that may detect Alzheimer’s through brain imaging, genetics and blood sample data.
If you could find out whether you’ll get Alzheimer’s disease years before symptoms like memory loss begin, would you want to know? What if it meant you could start making lifestyle changes that might buy you more time—or potentially help you avoid dementia altogether?
Canadian scientists said they have created an artificial intelligence program that can spot Alzheimer’s five years before symptoms set in.
Alzheimer’s is a challenging disease because once symptoms are noticeable, there are few changes a person can make that might limit the disease’s progression. But an early warning system like this A.I. program might help patients and family members make changes that could stretch out their cognitively healthy years—or at least give them time to discuss a plan for care. As long as there is no cure for Alzheimer’s and other dementias, prevention science that points to diet and exercise as modifiable risk factors is the best defense people have—and a test like this could give folks likely to develop Alzheimer’s the heads up they need to start making changes.
“At the moment, there are limited ways to treat Alzheimer’s and the best evidence we have is for prevention,” said Mallar Chakravarty, a neuroscientist at McGill University’s Department of Psychiatry. “Our A.I. methodology could have significant implications as a ‘doctor’s assistant’ that would help stream people onto the right pathway for treatment,” he said.
“For example, one could even initiate lifestyle changes that may delay the beginning stages of Alzheimer’s or even prevent it altogether,” said Chakravarty.
To design the algorithm that detects dementia, Chakravarty and his colleagues used data from MRIs, genetics and blood samples. The data was pulled from over 800 people who had normal cognition, early stages of memory impairment and Alzheimer’s disease. At this point, the team is confident in the accuracy of the algorithm—but more data from a bigger population would lead to even more accurate predictions.
“We are currently working on testing the accuracy of predictions using new data. It will help us to refine predictions and determine if we can predict even farther into the future,” says Chakravarty. With more data, the scientists would be able to better identify those in the population at greatest risk for cognitive decline leading to Alzheimer’s.
Currently, the only way for patients to get an accurate diagnosis is through testing using a PET scan or cerebrospinal fluid. Artificial intelligence could not only help identify those patients; it could also be a selection tool for participants for clinical trials. Some scientists attribute the colossal failure rate of dementia drugs to the possibility that drugs are tested on patients who already have Alzheimer’s and are advanced enough in the disease that the drugs fail to make a difference. The thinking now is that drugs should be tested on those in the early stages of the disease—but that’s a problem, considering that scientists have discovered Alzheimer’s could be decades in the making before symptoms show up.
The difficulty in finding clinical trial participants could be alleviated by a test like this one, and it could get patients who are at the greatest risk of developing dementia into clinical trials that might slow or stop the progression of the disease.
Artificial intelligence joins a plethora of other early detection tests—eye tests, blood tests, smell tests and genetic tests—that researchers are developing in order to safely and quickly evaluate who is most at risk for Alzheimer’s and who would most benefit from lifestyle prevention and clinical trial participation.
This study was published in the journal PLOS Computational Biology.