Analysis of task-based functional MRI data preprocessed with fMRIPrep

Abstract

Functional magnetic resonance imaging (fMRI) is a standard tool to investigate the neural correlates of cognition. fMRI noninvasively measures brain activity, allowing identification of patterns evoked by tasks performed during scanning. Despite the long history of this technique, the idiosyncrasies of each dataset have led to the use of ad-hoc preprocessing protocols customized for nearly every different study. This approach is time consuming, error prone and unsuitable for combining datasets from many sources. Here we showcase fMRIPrep (http://fmriprep.org), a robust tool to prepare human fMRI data for statistical analysis. This software instrument addresses the reproducibility concerns of the established protocols for fMRI preprocessing. By leveraging the Brain Imaging Data Structure to standardize both the input datasets (MRI data as stored by the scanner) and the outputs (data ready for modeling and analysis), fMRIPrep is capable of preprocessing a diversity of datasets without manual intervention. In support of the growing popularity of fMRIPrep, this protocol describes how to integrate the tool in a task-based fMRI investigation workflow.

Fig. 1 |. Overall workflow of the fMRIPrep protocol.

The analytic workflow is subdivided into three principal stages. First, a BIDS-compliant dataset is generated and validated. Next, dataset quality is assessed, and the data are preprocessed. Finally, the preprocessed data undergo a GLM fitting, which yields participant- and group-level statistical maps of task-related BOLD activity. FSL, FMRIB Software Library.

Fig. 2 |. Running fMRIPrep on HPC.

Execution of BIDS apps (such as MRIQC or fMRIPrep) is easy to configure on HPC clusters. This figure provides an example execution script for our SLURM-based cluster, Stanford’s Sherlock. An up-to-date, complete version of the script is available at https://fmriprep.readthedocs.io/en/latest/singularity.html#running-singularity-on-a-slurm-system.

Fig. 3 |. Output of the first-level analysis step.

Glass brain visualization of a statistical (z-stat) map reflecting the ‘intask-vs.-nontask’ activation obtained for subject 10 (thresholded at z = ±5).

Fig. 4 |. Group analysis results.

Visualization of a group statistical (z-stat) map (cluster threshold = 3.2) reflecting the ‘task-vs.-nontask’ activation for all subjects. a, Glass brain visualization. b, Left and right hemisphere surface plot visualizations. Visualizations were generated using a function from Nilearn (see Visualization of results for details).