A spatio-temporal reference model of the aging brain

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Gerontology

Background:

• Brain morphological changes are common in normal aging and neurodegenerative diseases like Alzheimer's (AD).
• Distinguishing these changes visually from MRI scans is challenging, hindering early diagnosis.
• A quantitative model of brain aging is needed to aid differential diagnosis.

Purpose of the Study:

• To develop a spatio-temporal model of brain morphological changes associated with normal aging.
• To create a reference distribution of brain morphology as a function of age.
• To enable the assessment of accelerated brain aging at population and individual levels.

Main Methods:

• Utilized groupwise image registration to analyze morphological variation across different age groups.
• Employed partial least squares regression to identify age-correlated deformation modes.
• Fitted smooth percentile curves to age-scores to establish a normative aging trajectory.

Main Results:

• The framework successfully extracted expected patterns of brain atrophy.
• Morphology scores were significantly lower in cognitively normal subjects compared to AD patients.
• Results suggest that accelerated aging contributes to morphological differences observed in AD.

Conclusions:

• The developed method effectively models spatio-temporal brain aging using MRI data.
• The model can identify accelerated brain aging, differentiating it from normal aging processes.
• This approach aids in the diagnosis of neurodegenerative disorders by providing a quantitative measure of brain aging.