The scripts in this repository perform analyses reported in the MEDIC paper. MEDIC itself is implemented in the Dosenbach and Greene Labs' processing pipeline. In order to run these analyses, you must have
There are several hard coded paths that may prevent you from running some scripts from this repo immediately. I will make an attempt to fix these over time. If you find something not working, feel free to create an issue!
To use this repo, first install it:
pip install -e ./ -v --config-settings editable_mode=strict
this will give you several scripts to run on your PATH.
Once you've installed this repo, you can run any of the scripts in the medic_analysis/scripts folder.
To call them, simply type the name of the script without the .py extension.
The two main scripts used in the paper are:
alignment_metrics.py: Generates alignment metrics for figures 5 and 6 in paper.paper_figures.py: Generates actual paper figures + supplemental figures.
Other scripts:
cerebellum_surface.py: Generates cerebellar surface from cerebellar volume segmentation.head_position_processing.py: Processes head position data for generating B0 change videos.resp_processing: Process respiratory belt/imaging data for looking at respiratory effects on B0.tSNR_processing.py: Does tSNR comparision between MEDIC and TOPUP data.
Note that this repo expects outputs from our lab processing pipeline.
For scripts written in MATLAB, you will need to navigate to the matlab_scripts directory and run them from there.
There are two main scripts, as well as an auxiliary Python script that goes along these scripts:
explore_MEDIC.m: This will run the ABCD group average comparison found in Figure 3 of the paper.split_groups.py: This Python script will split the ABCDavg dscalar into MEDIC and TOPUP groups. You must runexplore_MEDIC.mbeforehand.cluster_correct.m: This will run the multiple comparisons cluster correction for the t-statistic map. You should have runsplit_groups.pybeforehand.