Assessing distinct patterns of cognitive aging using tissue-specific brain age prediction based on diffusion tensor...

Area of Science:

• Neuroimaging
• Computational Neuroscience
• Biostatistics

Background:

• Multimodal brain imaging offers sensitive brain architecture and integrity measures.
• High-dimensional imaging data presents analytical challenges.
• Multivariate age prediction simplifies data into a single, biologically relevant measure.

Purpose of the Study:

• To explore differential brain age models using tissue-specific classifiers.
• To disentangle independent sources of heterogeneity in brain biology.
• To assess the cognitive sensitivity of distinct brain tissue classes.

Main Methods:

• Trained machine learning models to estimate brain age using FreeSurfer morphometry and diffusion tensor imaging (DTI) indices.
• Utilized data from 612 healthy controls (aged 18-87 years).
• Validated 11 models in an independent sample of 265 individuals (aged 20-88 years).

Main Results:

• Comprehensive morphometry and white matter microstructure models showed high accuracy in age prediction (r=0.83 and r=0.79, respectively).
• Deviations from chronological age were associated with cognitive performance on tests like spatial Stroop and symbol coding.
• Individuals with an over-estimated brain age exhibited poorer cognitive performance.

Conclusions:

• Tissue-specific brain age models provide sensitive measures of brain integrity.
• These models have implications for studying various brain disorders.
• Differential brain age trajectories and their cognitive correlates are characterized.