PEMAS Pipeline

Pipeline for Extricating Movement from Auricular Stimules (PEMAS). This repository contains code and scripts for analyzing 7T fMRI data acquired during a Respiratory-gated Auricular Vagal Afferent Nerve Stimulation (RAVANS) task and resting-state scans. The pipeline uses tools from Slocco's Laboratory, fMRIPrep, and FSL to preprocess data, regress physiological noise, and perform statistical analysis.

Data and Preprocessing

• Data Source: 7T fMRI data collected during RAVANS task and resting-state. Physiological recordings (heartbeat and respiration) were acquired for noise regression.
• Pipeline Foundation: The code is based on a pipeline provided by Slocco's Laboratory ((Lizbeth Ayoub, Andrew Bolender, and Roberta Slocco)).
• Physiological Noise Correction:
  • phycorr (RETROICOR and R-DECO) will be used to regress physiological noise. The physiological were collected using four channels in LabChart: heartrate, respiration, stimulus triggers and MRI triggers.
• fMRIPrep:
  • fMRIPrep will be used for comprehensive preprocessing, including:
    • Motion correction using aCompCor.
    • Slice-timing correction.
    • Spatial normalization to MNI and T1w space.
    • FreeSurfer recon-all for surface reconstruction (if enabled).
    • Generation of motion regressors from aCompCor.
• Output: Preprocessed BOLD data and motion parameter tables from aCompCor.

Analysis Steps

1. Brainstem and Medulla Mask Extraction:
   • 1_apply_brainmask_afterfmreiprep.sh extracts brainstem and medulla masks from the fMRIPrep output.
2. Motion Regressor Preparation:
   • 2_FD_FEAT_headmotion.sh processes the aCompCor motion parameter tables to generate text files containing regressors for FSL's FEAT.
3. Stimulus Extraction:
   • stem_trigger.m extracts stimulus timing information for FEAT input.
4. FEAT First-Level Analysis:
   • FSL's FEAT is used for first-level analysis, incorporating the preprocessed BOLD data, motion regressors, and stimulus timing.
5. FEAT Output Fusion:
   • GUTBRAIN_FLOBS_sumstat.m fuses the parameter estimates (PEs) from the FEAT first-level analysis.
6. Second-Level Analysis:
   • 3_fslmaths.sh calculates the difference between task and resting-state contrasts.
   • 5_fslmerge.sh merges the resulting contrast parameter estimate (COPE) images.
   • Randomise: FSL's randomise is then used for non-parametric permutation testing to perform statistical inference. Randomise allows for robust statistical testing, especially with data that may not meet the assumptions of parametric tests, by creating a null distribution through permutations of the data. This provides a way to control for multiple comparisons and determine the significance of observed effects.

Requirements

• FSL
• fMRIPrep
• MATLAB
• phycorr
• Bash

Proprietary Mentions

• This project builds upon base code and functions provided by the Slocco Laboratory at Spaulding Rehabilitation Hospital (Lizbeth Ayoub, Andrew Bolender, and Roberta Slocco).

Usage

Follow the numbered scripts and MATLAB functions in sequence to execute the analysis pipeline. Ensure that the necessary software and dependencies are installed and configured correctly.