Charting Multi-Scale Brain Phenotypes Using Spectral Normative Models
Date:
This talk at the ISMRM Workshop on 40 Years of Diffusion, was part of a session on AI and Big Data, during which I presented our recent advancements in developing spectral normative models for charting multi-scale brain phenotypes using diffusion MRI data. The talk highlighted how graph spectral brain eigenmodes can be leveraged to create scale-independent brain charts. We further explored the close association between spectral normative modeling and fundamental neural network architectures, such as graph neural networks (GNNs) and bayesian networks, discussing their implications for understanding brain structure and function across multiple spatial scales.
The slides for this talk are available here.