Add `process_data.sh`

[valosekj]

This PR adds the process_data.sh script. This script is intended to be the main script used within this repository.

Example usage:

# Clone this repo and go to it
# Then, run:
./process_data.sh \
-d ~/experiments/balgrist-sci/source_data/dir_20231010 \
-b ~/experiments/balgrist-sci/bids \
-r ~/experiments/balgrist-sci/data_processed \
-p sub-001 \
-s ses-01 \
-c T2w dwi \
-a 30 \
-x M

The process_data.sh script does the following steps:

• parse the input args
• activate the SCT conda environment with dcm2niix installed
• check if the necessary dependencies are installed (SCT, dcm2niix, fsleyes)
• convert DICOM files to NIfTI format using the file_loader.py script, which calls the dcm2niix function. The NIfTI images are stored in the folder specified by the -b arg.
• call the main analysis
  • rsync images from the bids folder (specified by the -b arg) to the results folder (specified by the -r arg)
  • run analysis -- so far only T2w SC seg

[valosekj]

@sandrinebedard, I think the PR is ready for review. Maybe you could test it on your new laptop to simulate the scenario that you are starting with no libraries installed :-D

[sandrinebedard]

I will test this on my new computer today!

[valosekj]

I am thinking that if the person wants to rerun the analysis, a reuse the already corrected manual seg, he won't be able since the dicom conversion needs to be done every time?

Good point! I've also been thinking about this scenario. I added an extra question asking the user whether to overwrite the folder in case it already exists in https://github.com/sct-pipeline/balgrist-sci/pull/5/commits/9330718390b13c0accd62c8dd5931cea81e91db8.

[valosekj]

Merging after an in-person discussion with @sandrinebedard.