Arian Ashourvan

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.

This paper uses HAVOK analysis to model nonlinear seizure dynamics and improves multi-day seizure risk forecasting by identifying short-period forcings.

This study tracks cortical activity propagations in youth, finding increased top-down propagations with cognitive control demands and development.

The paper shows that external inputs affect BOLD signal non-stationarity and presents a method to estimate these inputs using in silico models and fMRI data.

This paper introduces transparent graphene microelectrodes to simultaneously record microelectrocorticography and calcium fluorescence in vivo, revealing spatiotemporal seizure dynamics through non-negative matrix factorization.

The paper uses TMS-EEG and dynamic community detection to analyze how brain network reconfigurations differ in alpha and beta frequency bands following stimulation.

This paper introduces a dynamical systems framework to estimate both intrinsic brain activity and external stimuli, showing task-related changes and non-stationarities in cortical dynamics.

The paper uses a pairwise maximum entropy model to analyze intracranial EEG data and shows it captures neural dynamics and identifies white matter connections.

The paper uses machine learning to analyze brain states from fMRI data and shows that white matter connectivity influences their dynamics and development.

This study quantifies structure-function coupling between white matter connections and iEEG in epilepsy patients, finding increased coupling during seizures and its dependence on short-range structural connections.

This paper applies community detection algorithms to brain graphs derived from neuroimaging data to uncover patterns of coordinated neural activity and their implications for brain function.