SNAKE-fMRI: Why you should take T2* relaxation into account for simulating fMRI data

Motivation: Functional MRI (fMRI) helps study brain function but optimizing acquisition and reconstruction techniques for it is difficult due to a lack of reproducibility. Existing simulators are inflexible, hard to use with other tools, and slow. To address this issue, we created SNAKE-fMRI, an open-source simulator that can generate fMRI data. Goal(s): Investigate the impact of $$$T_2^*$$$ relaxation on image quality and statistical performance in fMRI simulation. Approach: We simulated two scenarios: EPI (Scenario 1) and stack-of-spirals (Scenario 2) acquisition. We also tested the impact of warm-start strategy, for reconstruction. Results: Neglecting $T_2^*$ relaxation can lead to misleading results for long readout. Impact: This work aims to provide a more realistic and reproducible framework for simulating fMRI data, leading to better methodology for the development of new acquisition and reconstruction strategies.