Alzheimer's Imaging Consortium.

Area of Science:

• Neuroimaging
• Neurodegeneration
• Brain Age Estimation

Background:

• Brain Age Gap (BAG) quantifies neurodegeneration by comparing brain MRI age to chronological age.
• BAG serves as a global measure to assess neurodegeneration differences across populations.
• This study investigates BAG in amyloid-positive individuals, focusing on sex differences within diagnostic groups.

Purpose of the Study:

• To apply a novel brain age prediction model to estimate Brain Age Gap (BAG) in Alzheimer's Disease Neuroimaging Initiative (ADNI) samples.
• To investigate potential sex differences in BAG among amyloid-positive individuals across various diagnostic categories.
• To explore the relationship between BAG, amyloid status, cognitive scores, and sex.

Main Methods:

• An ensemble model was developed using UK Biobank (UKBB) and Mayo Clinic Study of Aging (MCSA) data for brain age prediction from T1w MRI.
• The model employed robust preprocessing, data augmentation, and regularization for generalization.
• The trained model was applied to ADNI samples to estimate BAG without fine-tuning.

Main Results:

• No significant sex difference in BAG was observed in cognitively normal subjects, indicating model neutrality.
• BAG increased along the Alzheimer's Disease (AD) trajectory, consistent with prior research.
• After adjusting for age and ADAS13 scores, amyloid-positive females showed significantly older-appearing brains than males across most diagnostic groups (p < 0.05).

Conclusions:

• Female brains appear older than male brains in most diagnostic groups when considering amyloid positivity and cognitive scores.
• This finding suggests a potential difference in cognitive reserve between sexes, with females potentially having greater reserve.
• The study highlights sex-specific patterns in brain aging within neurodegenerative disease contexts.