AI maps local brain aging

Researchers at the University of Southern California, led by Andrei Irimia, developed a deep-learning AI model that generates detailed maps of local brain aging at the voxel level, rather than a single brain-age estimate. The model was trained on MRI scans from 14,748 cognitively normal adults aged 19-100 from six public datasets, including UK Biobank and Human Connectome Project, and tested on over 1,900 scans from the Alzheimer s Disease Data Initiative. The study, published in Proceedings of the National Academy of Sciences of the United States of America, found that frontal and temporal lobes appear biologically older than parietal and occipital regions, and the right hemisphere ages slightly faster than the left. In people with mild cognitive impairment or Alzheimer's disease, accelerated aging was observed in the hippocampus, amygdala, and other deep brain regions. Older local brain age correlated with poorer cognitive performance, especially in Alzheimer's patients. The model is a research tool, not yet a clinical diagnostic, and requires further validation with diverse datasets and longitudinal studies. The research was supported by the United States — National Institutes of Health and other funders.