A new study introduces a graph neural network (GNN) that can estimate the biological age of specific brain regions from MRI scans, providing a detailed map of how Alzheimer's disease (AD) accelerates aging across the cortex. The model, trained on over 14,000 cognitively normal adults, identifies which cortical features are most sensitive to AD pathology and links regional aging gaps to cognitive decline.
The Research
Led by Samuel D. Anderson and colleagues from the Alzheimer's Disease Neuroimaging Initiative, the study published on arXiv (January 2026) used T1-weighted MRIs of 14,423 cognitively normal adults to train a GNN that predicts local brain age (LBA) at high spatial resolution (mean inter-vertex distance = 1.37 mm). The model incorporates five morphometric features: cortical thickness, surface area, curvature, gray/white matter intensity ratio (GWR), and sulcal depth. On the ADNI dataset, the GNN outperformed existing state-of-the-art methods in mean absolute error and revealed biologically plausible aging patterns. In cognitively normal individuals, association cortices showed the most aging. In mild cognitive impairment, widespread aging was pronounced in the parahippocampal gyrus. Alzheimer's patients exhibited significant aging across the entire cortex, especially in medial temporal regions. Feature ablation identified curvature and GWR as particularly sensitive to AD pathology. Regional LBA gaps correlated significantly with neuropsychological measures of cognitive impairment.
Why It Matters
This research moves beyond average brain age to pinpoint where and how aging occurs at the individual level. For someone curious about their cognitive health, such local brain age maps could one day help detect early signs of neurodegeneration and track intervention effectiveness. The finding that curvature and gray/white matter ratio are especially sensitive to AD suggests these may be key biomarkers for monitoring disease progression.
What You Can Do
While this technique isn't available clinically, you can support your brain health with regular exercise, a balanced diet, mental stimulation, and social engagement. Stay curious about advances in neuroimaging—they're paving the way for personalized brain aging assessments.
Source: arXiv q-bio.NC
Curious about your own brain? Take our free adaptive IQ test or try 306 brain training levels.
