ComplexMultiEcho1

This project is part of a larger goal to improve fMRI data thorough removal of noise and artifacts. While the processing code here does not do everything, these data were collected with the goal of examining ways to combine:

• Thermal noise reduction methods that take advantage of using complex (vs magnitude-only) fMRI data and noise scans (building off of NORDIC)
• Multi-echo fMRI denoising (building off of tedana)
• Cardiac and respiratory noise removal using respiratory and cardiac recordings (potentially combining the AROMA approach with the tedana approach)

Data

• fMRI data with magnitude + phase + multi-echo, and noise scan
• Respiration data from a belt around the chest and pulse from a pulse oximeter
• Tasks
  • Visual vs Audio & Word vs Nonword task (3 8.5 min runs)
  • Movie viewing runs with different patterns of cued breathing (2-3 7.5 min runs)
  • Cued breathing "rest" runs where runs match the breathing patterns of a movie run (1-2 7.5 min runs)
  • DataCollectionGuide.md is guide/reminder for our data collection procedure

Organization & Processing Steps

1. SettingUpDirectories.md is instructions for getting the MRI, biopac, & psychopy data from a scanning session to where it belongs. This (will) include:

   • DICOM to BIDS with Personally identifiable information removed
   • PsychoPy run logs to GLM regressors for AFNI & BIDS (and eventually other relevant reporting)
   • TODO: Biopac traces to BIDS-style cardiac, respiratory time series & TTL pulse times.

2. AnatomicalProcessing.md is instructions for running freesurfer and generating the labelled ROIs-of-interest in EPI space.

3. FMRI_processing.md contains instructions for processing the fMRI data starting with afni_proc.py

4. Physiological_Proc.md have instructions for processing and analyzing the raw BIDS-converted respiration and cardiac traces.

Statistical Analysis

1. ISC_Group_Analyses.md show you how to perform Inter-subject correlations (ISC) on the Movie and Respiration runs.

2. Statistics.md will show how to extract the Visual-Audio FisherZ statistics, Degrees of Freedom (DOF), and Contrast-to-Noise Ratio (CNR) maps for the Visual-Audio and Word-NonWord conditions.

3. WNW_Group_Map_Generation.md shows you how to generate the final Vis-Aud and Word-Nonword group maps across the GLMs.

Visualization

1. Visualizations_for_Manuscript.md gives you instructions on how to re-generate all of the figures, visual QC checks, and where to locate the figures after they've been generated.