AI predicts dementia

AI-calculated brain age predicts dementia risk years before doctors can see it

August 20, 2026
9:51 pm

Memory clinics still rely on visual rating scales to assess brain atrophy, yet these tools lack the sensitivity to catch subtle changes associated with early dementia. A measure of how old a person's brain appears on brain scans compared with their actual age might be a potential predictor of who is more likely to develop dementia over the years, reports a new study published in Neurology.

Researchers in the Netherlands gathered medical data from patients enrolled in a national dementia or subjective cognitive impairment cohort between 2002 and 2024. Using advanced AI-based computer algorithms, they determined how old a person’s brain appeared on MRI scans and then compared that estimate with the person’s actual age.

They tested whether this mismatch, which they called brain-predicted age difference (brain-PAD), could help identify people with subjective cognitive decline (SCD) and mild cognitive impairment (MCI) who were more likely to develop dementia, and whether it added anything beyond what visual rating scales could already tell them.

AI-calculated brain age predicts dementia risk years before doctors can see it

Sign up for Blog Updates