Summary: Researchers have created a new tool that estimates an individual’s rate of aging from a single brain MRI. This brain-derived aging metric predicts future risk of chronic diseases and dementia years before symptoms appear. Unlike many prior “aging clocks,” the model was trained on longitudinal data from the same people across decades, reducing generational and environmental confounds.
The brain-based pace-of-aging measure correlates with worse cognitive performance, accelerated shrinkage of memory-related brain regions, and increased likelihood of developing dementia. Faster brain aging also associates with higher incidence of chronic conditions and greater mortality risk.
Key Facts:
- Early detection: A single structural MRI scan can estimate how quickly a brain is aging and flag elevated dementia risk well before symptoms.
- Body–brain link: Faster brain aging is linked to overall health decline, higher rates of chronic disease, frailty, and earlier death.
- Broad applicability: The tool performed consistently across diverse ethnic and socioeconomic groups, supporting wider use in research settings.
Source: Duke University
At many reunions we notice how differently people age. Some remain physically active and mentally sharp while others show signs of decline much earlier. “The way we age is distinct from our chronological age,” said Ahmad Hariri, professor of psychology and neuroscience at Duke University.
Researchers at Duke, Harvard and the University of Otago developed a freely available tool called DunedinPACNI (Dunedin Pace of Aging Calculated from NeuroImaging) that estimates an individual’s longitudinal pace of aging from a single brain MRI. Trained on repeated health assessments collected across decades, this approach captures the tempo of biological aging rather than differences that simply reflect generational exposures.
The model was trained using data from the Dunedin Study, a cohort of roughly 1,037 New Zealanders born between 1972 and 1973 who have been followed since birth. Repeated measurements—blood pressure, BMI, glucose and cholesterol, lung and kidney function, dental health and other markers—were used to compute each participant’s rate of aging over nearly 20 years. DunedinPACNI learned to predict that longitudinal pace of aging using only a single structural brain MRI collected at age 45 from 860 participants.
After training, the researchers validated the biomarker in multiple independent datasets from the U.K., U.S., Canada and Latin America. Across cohorts, faster DunedinPACNI scores were associated with lower cognitive test performance, faster hippocampal atrophy, earlier onset of memory and thinking problems, and a higher likelihood of conversion to clinically diagnosed dementia.
Faster brain aging predicts dementia and clinical decline
In one validation using scans from 624 participants aged 52 to 89 in a North American Alzheimer’s risk cohort, those identified as aging fastest by DunedinPACNI were about 60% more likely to develop dementia over subsequent years and experienced cognitive decline sooner than slower agers. The magnitude of the association surprised the team: “When we first saw the results, our jaws just dropped to the floor,” Hariri said.
Body and brain aging are connected
Beyond cognitive outcomes, higher DunedinPACNI scores predicted greater physical frailty, more age-related health problems (heart attacks, lung disease, stroke), and a higher near-term incidence of chronic disease. Participants with the fastest brain-aging profiles were estimated to be 18% more likely to receive a chronic disease diagnosis within several years and about 40% more likely to die within the same timeframe compared with average agers.
Importantly, the associations held across demographic groups not present in the original training set, including Latin American samples and U.K. participants who were low-income or non-White. “It seems to be capturing something reflected in all brains,” Hariri noted.
As global lifespans increase, so does the burden of age-related diseases like dementia. Better tools for monitoring biological aging could help identify individuals who might benefit most from early lifestyle or therapeutic interventions, and provide a way to evaluate whether anti-aging treatments slow the course of decline before irreversible brain damage occurs.
Although current Alzheimer’s treatments largely manage symptoms rather than reverse pathology, DunedinPACNI could help detect at-risk individuals earlier—when interventions might be more effective—and accelerate research into prevention strategies.
First author Ethan Whitman and colleagues emphasize that additional research is required to move DunedinPACNI from a research instrument to a clinical tool. For now, the team expects it to be particularly useful to researchers who have access to brain MRI data and aim to study aging processes and interventions. “We view it as a key new tool for forecasting disease risk and tracking progression, especially for Alzheimer’s and related dementias,” Hariri said.
Funding and disclosures: The authors have filed a patent application related to this work. Research support came from the U.S. National Institute on Aging (R01AG049789, R01AG032282, R01AG073207), the UK Medical Research Council (MR/X021149/1), and the New Zealand Health Research Council (Programme Grant 16-604).
About this neuroimaging and brain aging research news
Author: Robin Smith
Source: Duke University
Contact: Robin Smith – Duke University
Image: Image credited to Neuroscience News
Original Research (open access): “DunedinPACNI Estimates the Longitudinal Pace of Aging From a Single Brain Image to Track Health and Disease” by Ahmad Hariri et al., published in Nature Aging. The study describes the development and validation of DunedinPACNI as a next-generation brain MRI biomarker for tracking aging and predicting clinical outcomes.
Abstract
DunedinPACNI Estimates the Longitudinal Pace of Aging From a Single Brain Image to Track Health and Disease
To understand how aging contributes to functional decline and disease risk, accurate measures of how fast a person is aging are essential. Using the Dunedin Study, researchers developed DunedinPACNI—an MRI-derived measure that estimates longitudinal pace of aging from a single cross-sectional brain image. When applied to external datasets, faster DunedinPACNI predicted cognitive impairment, accelerated brain atrophy, conversion to diagnosed dementia, physical frailty, poorer overall health, future chronic disease, and mortality. Compared with the brain age gap metric, DunedinPACNI showed similar or stronger relationships with clinical outcomes. As a next-generation brain MRI biomarker, DunedinPACNI can help researchers study aging effects on health and evaluate anti-aging strategies.