Exporting volumetric ultrasound to DICOM

To follow up on this, @koeglfryderyk shared a sample phantom scan (can that be shared publicly here in this thread, Fryderyk?), which @David_Clunie reviewed and provided instructions, sample and conversion tool from Pixelmed toolkit. Can you please let us know Fryderyk if this issue can be resolved? It would be great if we could link the public phantom dataset and David’s instructions here for the benefit of the community.

1 Like

Hi all, just picked up this thread while looking into Dicom US options.

I would be interested to know if there was any followup on the use of Pixelmed for this particular use case

I think this can be resolved, but I’ll write another comment with an explanation of how to do it. Could you just tell me what is the best way of sharing files? (the NRRD has 30MB, the DICOM 100MB)

Volumetric ultrasound can be exported to DICOM using David Clunie’s tool com.pixelmed.convert.NRRDToDicom .

The documentation is here

To use it follow those steps:

  1. Make sure you have java installed, if not, install it from here
  2. Download pixelmed.jar
  3. run the following command in your terminal:
java -cp "path_to_pixelmed.jar" -Djava.awt.headless=true \
	com.pixelmed.convert.NRRDToDicom \
	"path_to_the_input.nrrd" \
	"path_to_the_output.dcm \
	"Patient Name" "Patient ID" "Study ID" "Series Number" "Instance Number"


A US phantom can be found here (in NRRD and exported to DICOM).

@koeglfryderyk I’ve just shared a dropbox folder with your gmail email - we can use that for sharing the files. Thank you for getting back to this discussion and answering the questions!

1 Like

I still have to figure out how to change the modality from OT to US after the export (David mentioned another one of his tools to do that, but I haven’t yet figured out how to do that. I’ll update the comment with the solution once I know how to do that

hello all, and really nice to see people investing efforts on this.

We’re working on the space of 3D/4DUS as well, and will be very interesting to have such a converter.

@koeglfryderyk , are you not keeping .mha sequence files?

we don’t have .mha files. We’re working with the Brainlab Curve neuronavigation system which only gives us the reconstructed 3D US volume