time-varying-brain-connectivity

Time-varying directed brain connectivity is estimated using a sliding-window prediction correlation (SWpC) workflow with an embedded linear time-invariant model. This approach derives directional strength and window-wise duration from multi-modal neuroimaging or neural recordings.

Sliding-window prediction correlation (SWpC) is a technique that infers directional connectivity within moving windows over neural data. It embeds a linear time-invariant model to quantify information flow directionality and transfer duration across each window.

Dynamic directed connectivity analysis applies to both resting-state and task-based neuroimaging. It supports multi-modal recordings including fMRI, EEG, and LFP for clinical stratification, cognitive neuroscience, and multimodal validation.

Inferring directional information flow over time is best achieved by applying SWpC-based directed connectivity analysis. This workflow estimates time-varying directional interactions in brain networks and provides guidance for interpretation and validation.

The SWpC-based directed connectivity workflow operates without specific external dependencies. You apply the analysis directly to your neuroimaging dataset to estimate time-varying information flow and derive directional strength.

Time-varying directed connectivity requires an embedded linear time-invariant model to quantify directionality and transfer duration accurately. The LTI model operates within each sliding window to derive directional strength for dynamic information flow.