The main objective of this interdisciplinary proposal is to integrate machine learning models with multimodal imaging, cognitive and neuropsychological data, using them to predict the brain age gap in healthy subjects and patients with neurodegenerative diseases (Alzheimer’s disease and frontotemporal dementia). We will build and validate predictive models and assess the synergy between imaging modalities, correlating brain age with cognitive decline and social determinants of health. Adopting a novel framework based on methods from artificial intelligence and computational biophysical simulations, we will extend brain age estimation towards data-driven multidimensional markers capable of mapping heterogeneous brain aging trajectories, applying them for the specific characterization of neurodegenerative disease sub-types. Finally, the neural mechanisms behind our models will be investigated with whole-brain biophysical simulations, allowing the in silico exploration of possible interventions to steer aging trajectories towards those of healthy subjects.
PI: Enzo Tagliazucchi
Support: ANID/Fondecyt
