Using data from 70 per cent of the participants, Liu’s team trained an artificial intelligence model to predict how old the participants were based on features of the brain images, such as the size of different brain regions and how distinct parts connected to each other. When the model was applied to the remaining 30 per cent of participants, its predictions were accurate to within 2.7 years of their actual ages.

Next, the researchers used the model to predict the age of a separate group of nearly 4700 people, aged 63 on average, who also had their brains imaged for the UK Biobank. The team calculated the difference between these participants’ actual ages and the ones predicted by AI, called the brain age gap. “The higher the AI-predicted age is relative to their actual age, the faster their brain is ageing,” says Liu.

https://www.newscientist.com/article/2459537-thirteen-proteins-in-your-blood-could-reveal-the-age-of-your-brain/