Scientists at Imperial College London used MRI scans and algorithms to produce computer-generated brain age and spot risk of dying young
Doctors may be able to warn patients if they are at risk of early death by analysing their brains, British scientists have discovered.
Those whose brains appeared older than their true age were more likely to die early and to be in worse physical and mental health, a study by Imperial College London found.
The research found a way of predicting someones brain age that could help to spot those at risk of dying young.
The study, piloted in Scotland, suggests using MRI scans to estimate a persons brain age compared with their real age could also help to spot who might be at increased risk of poor health as they grow older.
By combining MRI scans with machine learning algorithms, a team of neuroscientists trained computers to predict the age of a persons brain based on their volume of brain tissue.
When the technique was tested on a group of older adults in Scotland, they found that the greater the difference between the computer-generated brain age and the persons actual age, the higher their risk of poor mental and physical health and the more likely they were to die before they turned 80.
Those with a brain age older than their real age also had weaker grip, lower lung capacity and slower walking speed.
Researchers say that if the initial findings could be applied to a screening programme, the technique could be used to inform doctors, showing whether or not a patient had a healthy brain age or was above or below the line, similar to how body mass index (BMI) is used. They could then advise patients to change their lifestyle or start a course of treatment.
James Cole, a research associate who led the study, said: People use the age of an organ all the time to talk about health. Smokers are said to have lungs that are 20 years older than they should be, you can even answer online questionnaires about exercise and diet and get a heart age. This technique could eventually be like that.
However, it would need more fine-tuning for accuracy before it could be used in this way, Cole said. At present it has a margin of error of about five years. MRI scans are also currently too expensive to be used as a widespread screening tool but researchers hope that costs will come down in the future.
In the long run it would be great if we could do this accurately enough so that we could do it at an individual level, he said. However, at the moment, its not sufficiently accurate to be used at that sort of individual level.