Standardizing data acquisition with ReproIn¶
ReproIn¶
Program cards should be set up according to the ReproIn naming convention. Formatting your program card according to ReproIn allows the raw DICOM images to be automatically converted to BIDS-formatted NIfTI images using HeuDiConv.
Minimal example¶
$ tree
└── NormaL/NormaL_McDevitt/mystudy
└── mystudy
└── anat_scout
└── anat_T1w
└── func_task-localizer_run-1
└── func_task-localizer_run-2
└── fmap_dir-AP
└── fmap_dir-PA
Extended example¶
$ tree
└── NormaL/NormaL_McDevitt/mystudy
└── mystudy_ses-1
└── anat_ses-1_scout
└── anat_ses-1_T1w
└── func_ses-1_task-localizer_run-1
└── func_ses-1_task-localizer_run-2
└── fmap_ses-1_dir-AP
└── fmap_ses-1_dir-PA
└── mystudy_ses-2
└── anat_ses-2_scout
└── anat_ses-2_T1w
└── func_ses-2_task-localizer_run-1
└── func_ses-2_task-localizer_run-2
└── fmap_ses-2_dir-AP
└── fmap_ses-2_dir-PA
└── mystudy_ses-3
└── anat_ses-3_scout
└── anat_ses-3_T1w
└── func_ses-3_task-localizer_run-1
└── func_ses-3_task-localizer_run-2
└── fmap_ses-3_dir-AP
└── fmap_ses-3_dir-PA
Common mistakes¶
You should not have more than one scan with the exact same name (unless you stop and re-run a scan and you want to delete the first instance of that scan). When you have multiple scans with the exact same name, HeuDiConv will automatically add the BIDS-noncompliant _dup_ flag to the filename (indicating a duplicate run that can be deleted). In the example above, it is important that if you have multiple runs of the same task (func_ses-02_task-posttest) that you include _run-1, _run-2, etc. in the name. Make sure not to switch underscores and hyphens; underscores separate entities (e.g., key–value pairs like “task-study” and “run-1”), where as hyphens separate the suffix and extension of each key–value pair (e.g., “run” and “1”).
Resources¶
The BIDS Specification https://bids-specification.readthedocs.io/en/stable/02-common-principles.html
ReproIn walkthrough on Siemens Prisma console https://github.com/ReproNim/reproin/blob/master/docs/walkthrough-1.md