Title: Feature Attention Graph Neural Network for Brain Age Prediction
Speaker: Hae Sol Moon
Abstract:
Alzheimer’s disease (AD), a prevalent neurodegenerative disorder, is influenced by an intricate mix of risk factors including age, genetics, and environmental variables. In our study, we controlled genetic and environmental risks such as high-fat diet in mouse models. We utilized a Feature Attention Graph Neural Network (FAGNN), integrating various related datasets including brain structural connectomes, genetic traits, environmental factors, and behavioral data, to estimate “brain age” which is a quantitative brain health assessment metric. Our method demonstrated improved accuracy in age prediction over other methods and highlighted important age-associated brain connections with a distinct quadrant attention module. The most significant connections included the cingulum, striatum, corpus callosum, and hippocampus. We further investigated these findings through fractional anisotropy in different age groups of mice which underlined the significance of white matter degradation in the aging process. Our research underscores the effectiveness of integrative neural networks system in predicting brain age, and at the same time, finding important neural connections for brain aging.
