Dear Chris,

I am trying to convert diffusion data from dicom to nifti and I receive a warning message:

Warning: Slice timing appears corrupted (range 0..11487.5, TR=900 ms)

The output files I receive included .bvec, .bval, .json and .nii

So it seems to be working... but do I have to change or do something due to the warning message I receive.

Thank you very much for your help!

Best regards,
Hamza

This question is underspecified to give a good answer. Can you include a copy of the JSON file that will reveal the dcm2niix version used (e.g. v1.0.20201102), the manufacturer (e.g. Siemens) and the model (e.g. Vida XA10).

Based on incomplete information, one possibility is that this data Siemens XA data saved as classic (2D) format. If this is the case, I would suggest exporting the data from the MRI console in enhanced (4D) format to preserve meta-information. 

Regardless, I would be wary of using any tool that attempts to interpret the slice timing tags from the JSON file.

Dear Chris,

thank you for your quick answer! Attached is the JSON file. The data I am using is from an online database and so I cant export the data from the MRI console. Therefore I only have the DICOM files to work with.

Best regards,
Hamzah

Something is wrong with the DICOM header for these images. I would contact your Siemens Research Collaboration manager to resolve this. You may want to look at images copied direct from the scanner to see if an anonymization script or pACS is corrupting the CSA header. The TR is implausibly short, the bandwidth implausibly long, and the slice timing incompatible with the TR. 

RepetitionTime": 0.9,
"BandwidthPerPixelPhaseEncode": 24.284,
"SliceTiming": [
0,
1.03,
...
11.2125,