# Diffusion brain: DICOM to NRRD conversion

**URL:** <https://discourse.slicer.org/t/diffusion-brain-dicom-to-nrrd-conversion/19129>\
**Category:** Support\
**Created:** [August 9, 2021, 11:21pm UTC](https://discourse.slicer.org/t/diffusion-brain-dicom-to-nrrd-conversion/19129 "2021-08-09T23:21:00Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![rahulpaul](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/rahulpaul/32/11941_2.png) [@rahulpaul](https://discourse.slicer.org/u/rahulpaul)\
**Post date:** [August 9, 2021, 11:21pm UTC](https://discourse.slicer.org/t/diffusion-brain-dicom-to-nrrd-conversion/19129/1 "2021-08-09T23:21:00Z")

</div>

Operating system: MacOS  
Slicer version: 4.11

I am having little difficulty in converting the Dicom images to NRRD using DWIConvert. I believe something is not happening correctly during conversion as after conversion I can’t find any DWI components to choose. I am uploading the 24 sequences of DWI in Dicom. I can also upload the Dicom header info if anyone requires.

 ![Screen Shot 2021-08-09 at 5.24.12 PM](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/2/9/29f58cedf43f08ac81f702b3e0279c62a4ce9e28.jpeg)

Any help will be really appreciated.

---

<div class="post-metadata">

**Author:** ![Chris\_Rorden](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/chris_rorden/32/4073_2.png) [@Chris\_Rorden](https://discourse.slicer.org/u/Chris_Rorden)\
**Post date:** [August 10, 2021, 12:56pm UTC](https://discourse.slicer.org/t/diffusion-brain-dicom-to-nrrd-conversion/19129/2 "2021-08-10T12:56:10Z")

</div>

You might want to choose the `Extension Manager` menu item from the `View` menu and install the SlicerDcm2nii module.@ihnorton can provide more details. This will give you access to the DCM2niixGUI module where you can select an input folder and apply.

 ![dcm2niix](https://us1.discourse-cdn.com/flex002/uploads/slicer/original/3X/0/8/081a3b869019d9933359023d51ac17a60cb8bad9.png)

There are two other options:

1. Use [dcm2niix](https://www.nitrc.org/plugins/mwiki/index.php/dcm2nii:MainPage) from the command line. Provide the `-e y` to export to NRRD (instead of NIfTI), so the minimal command line would be `dcm2niix -e y /path/to/DICOMs`
2. Get [MRIcroGL](https://www.nitrc.org/plugins/mwiki/index.php/mricrogl:MainPage) and choose `Convert DICOM to NIfTI` from the `Import` menu. This provides a nice graphical wrapper for the many options available with dcm2niix. The relevant one for you is to choose NRRD from the `Output Format` drop down.

If you continue having issues, you may want to provide details regarding the scanner manufacturer (e.g. Bruker, Canon, GE, Philips, Siemens, UIH), as they each use different methods to define [DTI](https://www.na-mic.org/wiki/NAMIC_Wiki:DTI:DICOM_for_DWI_and_DTI) parameters.

---

<div class="post-metadata">

**Author:** ![rahulpaul](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/rahulpaul/32/11941_2.png) [@rahulpaul](https://discourse.slicer.org/u/rahulpaul)\
**Post date:** [August 12, 2021, 11:44am UTC](https://discourse.slicer.org/t/diffusion-brain-dicom-to-nrrd-conversion/19129/3 "2021-08-12T11:44:04Z")

</div>

@Chris_Rorden Thank you for your response. But its not working still. I can post the dicom headers, if that helps.

---

<div class="post-metadata">

**Author:** ![rahulpaul](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/rahulpaul/32/11941_2.png) [@rahulpaul](https://discourse.slicer.org/u/rahulpaul)\
**Post date:** [August 18, 2021, 12:01pm UTC](https://discourse.slicer.org/t/diffusion-brain-dicom-to-nrrd-conversion/19129/4 "2021-08-18T12:01:39Z")

</div>

@Chris_Rorden I sent you a google drive link with my data.

---

<div class="post-metadata">

**Author:** ![Chris\_Rorden](https://sea2.discourse-cdn.com/flex002/user_avatar/discourse.slicer.org/chris_rorden/32/4073_2.png) [@Chris\_Rorden](https://discourse.slicer.org/u/Chris_Rorden)\
**Post date:** [August 18, 2021, 12:45pm UTC](https://discourse.slicer.org/t/diffusion-brain-dicom-to-nrrd-conversion/19129/5 "2021-08-18T12:45:25Z")

</div>

@rahulpaul your DICOM images are from a Siemens VB17 MRI system. Unfortunately, the images were overzealously anonymized, removing the crucial sequence details stored in the [CSA header](https://nipy.org/nibabel/dicom/siemens_csa.html). The gradient directions and many other parameters are stored in the [CSA header](https://github.com/rordenlab/dcm2niix/tree/master/Siemens). Without these details, it is impossible for any conversion tool to infer these properties. I suggest you check the provenance of these images. Images directly from the scanner will have the CSA header intact, and you need to chose an anonymization tool that preserves this. This is a limitation of your images, not the tools used to import DICOM data.