questions > Converting enhanced dicom format
Showing 1-5 of 5 posts
Oct 5, 2018 04:10 PM | Jason Craggs - University of Missouri
Converting enhanced dicom format
Greeting,
Radiology recently installed a Siemens Vida system. It apparently exports in "enhanced dicom" format. It doesn't work with any of my software. Does anyone know anything about this format?
Radiology recently installed a Siemens Vida system. It apparently exports in "enhanced dicom" format. It doesn't work with any of my software. Does anyone know anything about this format?
Oct 5, 2018 04:10 PM | Chris Rorden
RE: Converting enhanced dicom format
Recent versions of dcm2niix should handle the new Siemens XA format
created by the Vida. There are two caveats,
a.) The Siemens XA Mosaic format is deprecated, and removes many elements needed for analyses. The Siemens XA Mosaic format is unsuitable for scientific work - this is not an issue with the converters, rather the fact that these DICOMs do not contain critical information. It is crucial for Siemens XA users to use the standard 2D XA output. I do suggest you lobby your Siemens research collaboration manager to remove the Mosaic format from future versions of Siemens XA to avoid accidental loss of data (in previous generations, saving in Mosaic was often a good idea, so users may uses this format out of habit).
b.) Be warned that the Siemens XA format no longer includes the proprietary Siemens CSA header. The Siemens XA enhanced format is currently very stripped down, and therefore does not provide some of the details useful for scientists. The data should still be useable, but the BIDS files created during conversion will not include many details provided be the previous generation Siemens datasets (e.g. phase encoding polarity, slice timing, effective readout time are probably missing). You will need to work out these details from your MRI console or from the sequence PDFs. Again, this reflects the relatively impoverished nature of the current Siemens XA format, rather than an error in the conversion. I do suggest you lobby your Siemens research collaboration manager to include these parameters in future versions of Siemens XA.
1.) I would first try the latest stable release of dcm2niix (v1.0.20180622)
https://github.com/rordenlab/dcm2niix/re...
2.) If (1) fails, you can try the latest development release
2a.) If you use Unix, you should be able to compile dcm2niix with these commands
$ git clone https://github.com/rordenlab/dcm2niix.gi...
$ cd ./dcm2niix/console
$ make
$ ./dcm2niix
Chris Rorden's dcm2niiX version v1.0.20181005 GCC6.1.0 (64-bit MacOS)
...
2b.) If you have Windows, you can click on the green AppVeyor icon from the dcm2niix Github home page and click on the artefacts link - this will provide a compiled executable of the latest commit
3.) If (2) fails, post an issue on Github
a.) The Siemens XA Mosaic format is deprecated, and removes many elements needed for analyses. The Siemens XA Mosaic format is unsuitable for scientific work - this is not an issue with the converters, rather the fact that these DICOMs do not contain critical information. It is crucial for Siemens XA users to use the standard 2D XA output. I do suggest you lobby your Siemens research collaboration manager to remove the Mosaic format from future versions of Siemens XA to avoid accidental loss of data (in previous generations, saving in Mosaic was often a good idea, so users may uses this format out of habit).
b.) Be warned that the Siemens XA format no longer includes the proprietary Siemens CSA header. The Siemens XA enhanced format is currently very stripped down, and therefore does not provide some of the details useful for scientists. The data should still be useable, but the BIDS files created during conversion will not include many details provided be the previous generation Siemens datasets (e.g. phase encoding polarity, slice timing, effective readout time are probably missing). You will need to work out these details from your MRI console or from the sequence PDFs. Again, this reflects the relatively impoverished nature of the current Siemens XA format, rather than an error in the conversion. I do suggest you lobby your Siemens research collaboration manager to include these parameters in future versions of Siemens XA.
1.) I would first try the latest stable release of dcm2niix (v1.0.20180622)
https://github.com/rordenlab/dcm2niix/re...
2.) If (1) fails, you can try the latest development release
2a.) If you use Unix, you should be able to compile dcm2niix with these commands
$ git clone https://github.com/rordenlab/dcm2niix.gi...
$ cd ./dcm2niix/console
$ make
$ ./dcm2niix
Chris Rorden's dcm2niiX version v1.0.20181005 GCC6.1.0 (64-bit MacOS)
...
2b.) If you have Windows, you can click on the green AppVeyor icon from the dcm2niix Github home page and click on the artefacts link - this will provide a compiled executable of the latest commit
3.) If (2) fails, post an issue on Github
Oct 5, 2018 06:10 PM | Jason Craggs - University of Missouri
RE: Converting enhanced dicom format
Hi Chris,
Thanks for the quick response!
I sent your explanation to several members of our imaging group and administrators (e.g., the chair of radiology). Several people have asked for a simplified explanation. Can you help me explain the importance of what you told me, and its implications for researchers, in a way they would understand? Any help with this is greatly appreciated.
Cheers,
Jason
Thanks for the quick response!
I sent your explanation to several members of our imaging group and administrators (e.g., the chair of radiology). Several people have asked for a simplified explanation. Can you help me explain the importance of what you told me, and its implications for researchers, in a way they would understand? Any help with this is greatly appreciated.
Cheers,
Jason
Oct 5, 2018 07:10 PM | Jason Craggs - University of Missouri
RE: Converting enhanced dicom format
Originally posted by Chris Rorden:
FYI:
Option 3 didn't work and I've posted an issue.
3.) If (2) fails, post an issue on Github
FYI:
Option 3 didn't work and I've posted an issue.
Oct 5, 2018 10:10 PM | Chris Rorden
RE: Converting enhanced dicom format