FM-fMRI: Event Conditioned Flow Matching for Rest-to-Task fMRI Time-Series Synthesis
A new model called FM-fMRI has been proposed for synthesizing task-based fMRI time series from resting-state fMRI data. This event-conditioned flow-matching model generates task region of interest (ROI) time series by learning a continuous-time conditional vector field, enabling efficient sampling and flexible conditioning over various event schedules.
WPN Brief
- What Happened
A new model called FM-fMRI has been proposed for synthesizing task-based fMRI time series from resting-state fMRI data. This event-conditioned flow-matching model generates task region of interest (ROI) time series by learning a continuous-time conditional vector field, enabling efficient sampling and flexible conditioning over various event schedules.
- Why It Matters
The development of FM-fMRI is significant as it enhances the ability to analyze neural dynamics associated with specific tasks, potentially improving research in areas such as autism and advancing the understanding of brain connectivity through its strong spectral and connectivity agreement.