Based on the biological age of the brainnew artificial intelligence technology can accurately calculate a person’s risk of cognitive decline and Alzheimer’s disease.
The human brain may hold clues to the future course of our health, US scientists report. In a study they conducted, they found that your biological age can reveal how much risk a person is for developing neurodegenerative problems, such as Alzheimer’s disease.
In fact, they developed a new artificial intelligence model, which analyzes MRIs of the brain and accurately determines the deterioration of mental functions that can lead to disease in the future.
He The system detects subtle changes that indicate whether the brain is more “aged” than a person’s chronological age. It also identifies some invisible markers in the brain’s anatomy that correlate with cognitive decline.
«Our study harnessed the power of deep learning to identify areas of the brain that are aging in a way that suggests cognitive decline.“, said the principal investigator, Dr. Andrei Irimiaassistant professor of Gerontology and Biomedical Engineering at the University of Southern California (USC).
«People age at different rates. So are the deep tissues within their bodies. The brain of a 40-year-old can look like the brain of a 30-year-old on MRIs. Or it could be more like the brain of a 60 year old.«.
the new research
The new findings are published in the scientific journal Proceedings of the National Academy of Sciences. As the researchers explain, they collected brain MRIs from 4,681 volunteers, who at the time had normal mental function.
Over the next few decades, 351 of the participants experienced a decline in their cognitive functions. Another 359 developed Alzheimer’s disease.
The scientists created an artificial intelligence model to estimate the biological age their brains originally were. Specifically, they programmed it to use MRIs to generate detailed anatomical maps of the brain. These maps revealed the pattern of brain aging in each individual participant.
Based on these maps, determined the biological age of each participant’s brain. Then, they compared the biological age with the actual (chronological) age. The greater the difference between the two, the worse the mental performance of the participants turned out to be.
High accuracy
They then verified their findings by running the model on another 1,170 volunteers. Their results showed that their model could predict with an accuracy of 2.3 years the actual (chronological) age of each healthy participant. This precision is superior to that of other similar models that have been created.
The new research also showed that men’s brains age differently than women’s. In fact, some parts age faster in men and others in women.
«Our technology may be a powerful tool for estimating future risk of developing Alzheimer’s disease.said Dr. Irimia. “The sooner high-risk people are identified, the sooner interventions to protect them can begin.
Our model captures subtle and complex features of aging. These cannot be perceived by other methods. That is why he is so precise in his estimates.