Joaquín Goñi

This paper introduces a concept-driven disentanglement framework for interpretable GNNs to predict brain functional connectivity from structural connectivity, achieving high accuracy and providing interpretable insights.

The paper introduces a homological formalism to analyze higher-order properties of human brain functional sub-circuits and identifies task-specific homological differences in brain connectivity.

This paper introduces griPEER, a regularization method that uses structural brain connectivity to improve regression coefficient estimation in generalized linear models, and applies it to classify HIV+ and HIV- individuals.

This paper identifies two spatially independent functional connectivity components across EEG and fMRI data, revealing linked but distinct connectivity dynamics across timescales.

The paper introduces a regularization method for regression that incorporates brain connectivity to improve estimation of associations between brain structure features and scalar outcomes.