Raphaël Liégeois

The paper introduces an arrow-of-time sensitive metric to identify causal triggers of brain function in neuroimaging data, revealing more localized causal effects than traditional connectivity measures.

The paper embeds brain networks into high-dimensional graph spaces using nodal invariants and shows improved classification accuracy in higher dimensions.

The paper uses topological data analysis to reveal a hub-like transition state in resting brain activity, showing consistent dynamics within individuals and potential for precision medicine.

This paper uses graph signal processing to decode brain tasks and fingerprint individuals by analyzing structure-function coupling in brain imaging data.

This paper investigates the temporal dynamics of human brain fingerprints, finding that longer time scales (~300s) yield the best identification, while shorter bursts correlate with neuronal activity and specific cognitive functions.

The paper reviews null models for resting-state fMRI time series and discusses their interpretation, arguing they are useful but not mandatory for validating neuroimaging applications.

The paper applies dynamic mode decomposition to fMRI data to identify dynamic modes that capture spatio-temporal brain activity patterns, revealing new insights beyond static functional connectivity.

This study finds that global signal regression enhances the association between resting-state functional connectivity and behavioral measures in young healthy adults.

This paper examines statistical issues in dynamic functional connectivity analysis using resting-state fMRI data and suggests AR models may better capture temporal fluctuations than static FC or HMM approaches.