BIDS difficulties for a newbie NOT_INCLUDED

#1

Hi,

Sorry if the solution is really obvious. I’m getting some MRI data ready for pre-processing, and can’t get past these BIDS errors:-

(code: 1 - NOT_INCLUDED)
                ./sub-03/ses-1/anat/sub-03_ses-1_dir-cor_T1w.json
                        Evidence: sub-03_ses-1_dir-cor_T1w.json
                ./sub-03/ses-1/anat/sub-03_ses-1_dir-cor_T1w.nii
                        Evidence: sub-03_ses-1_dir-cor_T1w.nii
                ./sub-03/ses-1/anat/sub-03_ses-1_dir-tra_T1w.json
                        Evidence: sub-03_ses-1_dir-tra_T1w.json
                ./sub-03/ses-1/anat/sub-03_ses-1_dir-tra_T1w.nii
                        Evidence: sub-03_ses-1_dir-tra_T1w.nii
                ./sub-03/ses-1/func/sub-03_ses-1_bold.json
                        Evidence: sub-03_ses-1_bold.json
                ./sub-03/ses-1/func/sub-03_ses-1_bold.nii
                        Evidence: sub-03_ses-1_bold.nii
  • What’s wrong with sub-03_ses-1_bold?
  • How should I handle the different varietes of T1 I’ve got? I get “T1”, “T1_MPR_cor”, “T1_MPR_tra”, and “T1_S5_DIS2D”.

Thanks!
Alistair

#2

Hi @alistaircannon,

All of your nifti files should be compressed, so nii.gz instead of .nii

What you’re currently doing is using the dir entity label to denote the differences between these anatomical images. I would instead suggest using the acq entity label. So T1_MPR_cor can use acq-cor, and T1_MPR_tra can use acq-tra, and so on. I’m not entirely sure what all your anatomical varieties are, but BIDS will want your anatomical acquisitions to be 3D. Your scanner may generate acquisitions that aren’t BIDS-compliant, in which case you would simply exclude them from BIDS conversion.

Hope this helps.

#3

Hi Dan,

Amazing, thank you! That makes more sense. Changing dir-cor to acq-cor has satisfied the BIDS checker. I’ll exclude the S5_DIS2D.

I’ve compressed the BOLD nifti so it’s now sub-03_ses-1_bold.nii.gz. I still get a code: 1 - NOT_INCLUDED though:

                ./sub-03/ses-1/func/sub-03_ses-1_bold.json
                        Evidence: sub-03_ses-1_bold.json
                ./sub-03/ses-1/func/sub-03_ses-1_bold.nii.gz
                        Evidence: sub-03_ses-1_bold.nii.gz

Do you know why this might be?

Thanks again,
Alistair

#4

Sorry, I forgot to mention earlier that for functional acquisitions, the task entity label is required in the file names; this applies to both the .nii.gz and .json files. Additionally, in the functional JSON files, there needs to be a TaskName field with the task name (e.g. TaskName: bart)

If you’re not using a BIDS converter to help transform your data into a BIDS-compliant dataset, I’d suggest checking out this list.

#5

Brilliant, thank you. Huge success! :tada:

I used dcm2niix, then used a scrappy python script I wrote to rename the files. I’ll find a tool from that list, which will probably be much better.

Thanks again,
Alistair

#6

if you are specifically interested in BIDS conversion, you have this list: