Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Adding header information from NX VA10A Enhanced DICOMs #240

Closed
lucijj opened this issue Oct 19, 2018 · 34 comments
Closed

Adding header information from NX VA10A Enhanced DICOMs #240

lucijj opened this issue Oct 19, 2018 · 34 comments
Labels
enhancement

Comments

@lucijj
Copy link

lucijj commented Oct 19, 2018

We have recently started using dcm2niix to convert our Siemens Vida's enhanced DICOMs and generate BIDS/json files. The conversion works well, but there are many parameters present in the enhanced DICOM header that are not included in the nii header or the json files. In particular, the SharedFunctionGroupsSequence and PerFrameFunctionalGroupsSequence DICOM tags have a wealth of info; much of which was included and is still parsed from the Siemens CSA headers on N4 mosaic DICOMs. It would be very helpful to have those included in the BIDS/json files and the nii headers.

I'm requesting that the these extra data be included in the nii headers and the BIDS/json files. I would be happy to provide technical assistance, beta testing, and sample datasets if it would be helpful.
@neurolabusc
Copy link
Collaborator

  • I have received a few Vida datasets, but most have been destructively de-identified or saved in the mosaic or classic format that removes a lot of tags, others are datasets I can not share. A well curated, publicly shareable dataset saved in enhanced mode without de-identification would be great.
  • We should be able to extract Phase Encoding Polarity from 0021,111c
  • I agree there seems to be some data in the PerFrameFunctionalGroupsSequence (5200,9230), but it is unclear to me what what they all are. dicm2nii provides some guesses. If you want to provide a decoder-ring, we could enrich the BIDS data generated for XA10A.

@lucijj
Copy link
Author

lucijj commented Oct 19, 2018

Excellent! Yes, I have a pretty good map of the values. I'll attach the link below. I'll also get you a good dataset. The ones I have now are all form a project, and rather than try to get permission and clearance, I'll just scan a phantom. I assume you'll want a functional and a DW run. Anything else that might be helpful?

Thanks a ton!

https://utexas.box.com/s/1mpz77srx69wk2eqngxcwg3xd8iuq7me

@lucijj
Copy link
Author

lucijj commented Oct 20, 2018

OK, here are the corrected links. I have tested them this time, so they should work.

Link to freely shareable dataset with enhanced DICOMs: https://utexas.box.com/shared/static/krqu5zuurstqp29v15258awhgu5v5qq6.zip

Link to DICOM tag map of the first slice of the DTI wSMS=3 woGRAPPA run in that dataset: https://utexas.box.com/shared/static/b68p1yjc0wk0pbfu6m6i5xeo3u2k8eef.txt

Link to the PDF dump of acquisition parameters for the full protocol: https://utexas.box.com/shared/static/0vgjhps87tk2vlxtimc99g4tvrcvg3jr.pdf

neurolabusc added a commit that referenced this issue Oct 20, 2018
@neurolabusc
More XA10A Vida header info (#240)
32bdca6
@neurolabusc
Copy link
Collaborator

neurolabusc commented Oct 20, 2018 •
edited
Loading

Thanks for the example. The following header BIDS tags are generated for Vida data:

  • 0018,0091; 0018,9231; 0021,1009 -> PartialFourier
  • 0021,1104 -> SliceTiming, MultibandAccelerationFactor
  • 0021,111C -> PhaseEncodingDirectionPositive
  • 0021,1142 -> DwellTime
  • 0021,114f -> ReceiveCoilActiveElements
  • 0021,1156 -> ParallelReductionFactorInPlane
  • 0018,1250 -> CoilString

Thats about all I could decipher. As I do not have access to Vida XA10 hardware, I would ask @lucijj, @BrainStormCenter, @mitchem890, @captainnova to test this release and see if you can make any more suggestions.

The slice timing reported for the Siemens SMS product sequences is unusual, but does seem to match the tags. I have asked Siemens to validate the temporal slice timing used in these sequences.

Note you exported your data as both enhanced and mosaic images. Be aware that the Vida XA10 mosaic images are missing Type 1 information such as slice orientation. Therefore they are not valid DICOM data, they lack vital fields, and no conversion can fix those defects. Please train you users to avoid exporting as mosaics to avoid data loss.

@xiangruili
Copy link

I verified this one based on "Echo Spacing" in the PDF:
0021,1153 'FD' 'BandwidthPerPixelPhaseEncode'
This is useful to compute echo spacing (dwell time), while 0021,1142 -> RealDwellTime has different mean.

neurolabusc added a commit that referenced this issue Oct 21, 2018
@neurolabusc
XA10A Vida dwell time (#240)
d0470f1
@neurolabusc
Copy link
Collaborator

Thanks @xiangruili - update generates values that try to estimate same parameters as previous Siemens BIDS, delving with partial Fourier and iPAT settings. E.G.:

  • "EffectiveEchoSpacing": 0.000339997,
  • "DerivedVendorReportedEchoSpacing": 0.000679995,
  • "TotalReadoutTime": 0.0404597,

@lucijj
Copy link
Author

lucijj commented Oct 22, 2018 •
edited
Loading

I have acquired a sample dataset and scoped out the read gradient in order to independently validate the timing parameters. The link below includes the DICOMs, the scope data, and my processed scope data in MATLAB mat-files. My method and the results are written up (i.e. slice timing, echo spacing, total readout time) in the text file, writeup.txt.

I have intentionally not looked at the DICOMs yet. I wanted to be blinded as I worked up the gradient data in MATLAB. So, could someone please download them and work them up in dcm2niix to verify my measured timings? That would tie the knots on this nicely, I think.

https://utexas.box.com/shared/static/tuga6ov2iz20kephz0rr05fokfx521nj.zip

@neurolabusc neurolabusc mentioned this issue Oct 22, 2018
Closed
@iamdamion
Copy link

I am following along with this ticket and using the same data as @lucijj from the same scanner. To clarify, should the above (specifically I'm most concerned with TotalReadouttime) be showing up in the json with the newest files available today? Or is this still not pushed to master? Thanks!

neurolabusc added a commit that referenced this issue Oct 22, 2018
@neurolabusc
Vida partial Fourier update (#240)
0855b37
@neurolabusc
Copy link
Collaborator

Yes, if you run v1.0.20181022 and choose to save a BIDS file (-b y) you will get a slew of information.

Of specific note, be aware that the BIDS tag TotalReadOutTime should be called "EffectiveReadOutTime", but obeys the BIDS standard and the expectations of FSL. The file nii_dicom_batch.cpp includes more notes, as well as @mharms Excel file and sampe images. This explains why iPAT influences this value but partial Fourier does not.

If you run Windows, you can get a compiled version of the latest commit by clicking on the green AppVeyor icon on the project home page and then clicking on the artifacts button. For Unix, you can build a minimal version of the latest commit with

git clone https://github.com/rordenlab/dcm2niix.git
cd dcm2niix/console
make

Here is the BIDS file for series 4 of the timing study @lucijj uploaded.

{
	"Modality": "MR",
	"MagneticFieldStrength": 3,
	"Manufacturer": "Siemens",
	"ManufacturersModelName": "MAGNETOM_Vida",
	"InstitutionName": "University_of_Texas_at_Austin",
	"InstitutionAddress": "Trinity_1701_Austin_Houston_US_78712",
	"DeviceSerialNumber": "175689",
	"StationName": "AWP175689",
	"BodyPartExamined": "BRAIN",
	"PatientPosition": "HFS",
	"ProcedureStepDescription": "dcm2niix_Samples",
	"SoftwareVersions": "syngo_MR_XA10",
	"MRAcquisitionType": "2D",
	"SeriesDescription": "REST1_w_SMS_PF0",
	"ProtocolName": "REST1_w_SMS_PF0",
	"ImageType": ["ORIGINAL", "PRIMARY", "M", "NONE"],
	"SeriesNumber": 4,
	"AcquisitionTime": "08:32:16.845000",
	"AcquisitionNumber": 1,
	"SliceThickness": 2.4,
	"SpacingBetweenSlices": 2.4,
	"EchoTime": 0.03,
	"RepetitionTime": 0.5,
	"FlipAngle": 52,
	"PhaseEncodingPolarityGE": "Unflipped",
	"ReceiveCoilActiveElements": "HE1-4",
	"CoilString": "HeadNeck_20_TCS",
	"MultibandAccelerationFactor": 2,
	"PercentPhaseFOV": 100,
	"EchoTrainLength": 64,
	"AcquisitionMatrixPE": 64,
	"ReconMatrixPE": 64,
	"BandwidthPerPixelPhaseEncode": 24.414,
	"EffectiveEchoSpacing": 0.000640002,
	"DerivedVendorReportedEchoSpacing": 0.000640002,
	"TotalReadoutTime": 0.0403201,
	"PixelBandwidth": 2298,
	"PhaseEncodingDirection": "j-",
	"SliceTiming": [
		0.16,
		0,
		0.32,
		0.16,
		0,
		0.32	],
	"ImageOrientationPatientDICOM": [
		1,
		-5e-13,
		1e-06,
		-5e-13,
		1,
		1e-06	],
	"InPlanePhaseEncodingDirectionDICOM": "COL",
	"ConversionSoftware": "dcm2niix",
	"ConversionSoftwareVersion": "v1.0.20181022  GCC6.1.0"
}

@iamdamion
Copy link

Perfect! I got the same results on my end. Thanks!

@BrainStormCenter
Copy link

Greetings,

Chris has been helping me get our Vida data into something usable.
I notice some differences in output. For example, my output doesn’t contain the slice timing information. Does anyone have a suggestion why this might be?

Cheers,
Jason

-- Version information:
Chris Rorden's dcm2niiX version v1.0.20181005 Clang10.0.0 (64-bit MacOS)

-- Command
dcm2niix -b y -ba n -f %p_%s -o ../SP022dcm2niix/ ./

--terminal output
Found 704 DICOM file(s)
slices stacked despite varying acquisition numbers (if this is not desired recompile with 'mySegmentByAcq')
Convert 120 DICOM as ../SP022dcm2niix/fMRI_(Pain_1)21 (76x76x36x120)
slices stacked despite varying acquisition numbers (if this is not desired recompile with 'mySegmentByAcq')
Warning: Saving 98 DTI gradients. Validate vectors (image slice orientation not reported, e.g. 2001,100B).
Convert 98 DICOM as ../SP022dcm2niix/DTI_5 (244x244x32x98)
slices stacked despite varying acquisition numbers (if this is not desired recompile with 'mySegmentByAcq')
Convert 120 DICOM as ../SP022dcm2niix/fMRI(resting)17 (76x76x36x120)
slices stacked despite varying acquisition numbers (if this is not desired recompile with 'mySegmentByAcq')
Convert 120 DICOM as ../SP022dcm2niix/fMRI(resting)13 (76x76x36x120)
slices stacked despite varying acquisition numbers (if this is not desired recompile with 'mySegmentByAcq')
Convert 120 DICOM as ../SP022dcm2niix/fMRI(Pain_3)29 (76x76x36x120)
slices stacked despite varying acquisition numbers (if this is not desired recompile with 'mySegmentByAcq')
Convert 120 DICOM as ../SP022dcm2niix/fMRI(Pain_2)_25 (76x76x36x120)
Convert 1 DICOM as ../SP022dcm2niix/SAG_T2_SPACE_4 (256x256x176x1)
Convert 1 DICOM as ../SP022dcm2niix/DTI_7 (244x244x32x1)
Convert 1 DICOM as ../SP022dcm2niix/t1_space_sag_p2_iso_2 (256x256x192x1)
Convert 1 DICOM as ../SP022dcm2niix/gre_field_mapping_3_ph (64x64x36x1)
Convert 1 DICOM as ../SP022dcm2niix/DTI_8 (244x244x32x1)
Conversion required 63.934312 seconds (63.496799 for core code).

--Output in .json file
{
"Modality": "MR",
"MagneticFieldStrength": 3,
"Manufacturer": "Siemens",
"ManufacturersModelName": "MAGNETOM_Vida",
"InstitutionName": "University_Hospital_Clinics",
"InstitutionAddress": "Hospital_Dr_1_Columbia_Central_US_65201",
"DeviceSerialNumber": "175654",
"StationName": "AWP175654",
"SeriesInstanceUID": "1.3.12.2.1107.5.2.50.175654.2018101209404933144176809.0.0.0",
"StudyInstanceUID": "1.3.12.2.1107.5.2.50.175654.30000018101205284676700000058",
"PatientName": "SPO22_001",
"PatientID": "2018.10.12-09:06:49-STD-1.3.12.2.1107.5.99.3",
"PatientSex": "O",
"PatientWeight": 92.9864,
"BodyPartExamined": "HEAD",
"PatientPosition": "HFS",
"ProcedureStepDescription": "HEAD_ROUTINE_20_channel_coil",
"SoftwareVersions": "syngo_MR_XA10",
"MRAcquisitionType": "2D",
"SeriesDescription": "fMRI_(Pain_1)",
"ProtocolName": "fMRI_(Pain_1)",
"ImageType": ["ORIGINAL", "PRIMARY", "M", "NONE"],
"SeriesNumber": 21,
"AcquisitionTime": "09:40:58.365000",
"AcquisitionDateTime": "2018-10-12T09:40:58.365000",
"AcquisitionNumber": 1,
"SliceThickness": 3,
"SpacingBetweenSlices": 3,
"EchoTime": 0.03,
"RepetitionTime": 2.46,
"FlipAngle": 75,
"PercentPhaseFOV": 100,
"EchoTrainLength": 76,
"ReconMatrixPE": 76,
"PixelBandwidth": 1827,
"PhaseEncodingAxis": "j",
"ImageOrientationPatientDICOM": [
1,
-5e-13,
1e-06,
-5e-13,
1,
1e-06 ],
"InPlanePhaseEncodingDirectionDICOM": "COL",
"ConversionSoftware": "dcm2niix",
"ConversionSoftwareVersion": "v1.0.20181005 Clang10.0.0"
}

@lucijj
Copy link
Author

lucijj commented Oct 23, 2018 via email

@BrainStormCenter
Copy link

BrainStormCenter commented Oct 24, 2018 via email

@iamdamion
Copy link

@BrainStormCenter,

I followed the instructions above from @neurolabusc and they worked perfectly for me. Might be worth removing and re-installing following those instructions: #240 (comment)

I believe he's planning on releasing a full update release soon, so you could also wait for that if it's not urgent.

@captainnova
Copy link
Collaborator

captainnova commented Oct 24, 2018 via email

@mharms
Copy link
Collaborator

mharms commented Oct 24, 2018

Is someone willing to summarize after this recent flurry of development (1) which BIDS fields are not available in Enhanced Dicom from XA10 that were available in the Siemen's Dicoms prior to XA10, and (2) which, if any, of the BIDS fields that are now being extracted for XA10 still need to be treated cautiously?

@BrainStormCenter
Copy link

I’ve completely reinstalled dcm2niix and got the updated version. I really have to thank Chris for all the help and effort he's put into dealing with these Vida issues.

@BrainStormCenter
Copy link

BrainStormCenter commented Oct 24, 2018 •
edited
Loading

Example comparison of Siemens Vida (syngo_MR_XA10) and TrioTim (syngo_MR_B17) dicom headers.
Converted with dcm2niix.
Version: v1.0.20181022 Clang10.0.0

The extensions are changed from json to txt for uploading.

TrimTrio_dicom.txt ('un'enhanced ) = dcm2niix converted dicom files from the Siemens TrimTio
Vida_ehn_dicom.txt (enhanced) = dcm2niix converted enhanced-dicom files from the Siemens Vida

TrioTim_dicom.txt
Vida_ehn_dicom.txt

@mharms
Copy link
Collaborator

mharms commented Oct 24, 2018

@BrainStormCenter Could you run the conversion of the Trio data with the same version of dcm2niix? There are a bunch of fields missing in your Trio example, solely because it was converted with a different (greater than year older) version of dcm2niix. Also, do you by any chance have Vida enhanced Dicoms for either dMRI or fMRI acquired using Siemens SMS/MB? If so, it would be good the see the json for that as well (to see if all the MB related stuff is getting reported).

@neurolabusc It seems a bit odd that the Vida json should have a "PhaseEncodingPolarityGE" flag. Should that tag perhaps be renamed?

@BrainStormCenter
Copy link

@mharms You asked a completely reasonable question, I should have thought of that before posting. I’ve updated the files. Regarding your other questions, the answer is maybe... 🤔 I just don't know exactly what you're asking for so I don't know. Sorry. What do you mean by SMS/MB and MB related stuff?
Cheers,
Jason

@mharms
Copy link
Collaborator

mharms commented Oct 25, 2018

Thanks. So, @neurolabusc, are ScanningSequence, SequenceVariant, ScanOptions, and SequenceName (which I think are all standard Dicom fields; i.e., not part of the CSA header) not included in the Vida Enhanced Dicoms?

@BrainStormCenter SMS/MB = Simultaneous MultiSlice/ Multiband; I believe that Vida XA has Siemens product SMS implementation available (assuming you have the SMS license), so I was wondering if you could do an acquisition with MB and post the json from that?

@neurolabusc
Copy link
Collaborator

neurolabusc commented Oct 25, 2018 •
edited
Loading

strok- @mharms, you are correct, I should probably remove the PhaseEncodingPolarityGE tag from the general release. It reflects the value for the Vida tag PhaseEncodingDirectionPositive (0021,111C) - this should be redundant with the sign of the field PhaseEncodingDirection. However, I have not yet seen Vida data where phase encoding polarity has been reversed. At the moment I assume 0021,111C is either 0 or 1, but it would be nice to have a sample dataset to confirm.

  • @mharms which, if any, of the BIDS fields that are now being extracted for XA10 still need to be treated cautiously. All the values seem reasonable based on the limited datasets I have seen. As noted, I have yet to see any images with reversed phase encoding polarity, so it would be good to validate PhaseEncodingDirection. I also can not replicate the typo reported by @captainnova - so it would be nice to track down what is going on there. Beyond that, it would be nice to have more eyes look at the values and more sample datasets.

  • @mharms - the dataset provided by @lucijj does include both fMRI and DTI with SMS/MB. We detect this with TimeAfterStart (0021,1104, in seconds). We could also do this with FrameAcquisitionDateTime (0018,9074 in DT format) but this tag is scrambled by Siemen's in-built anonymization so I consider it unreliable (while 0021,1104 is removed by de-identification: I would rather report no value than incorrect values). Both 0021,1104 and 0018,9074 provide the same plausible solutions for slice timing and multiband after the first volume, and both have been observed to provide the wrong solution for the first volume. Hence dcm2niix only reports slice timing and multiband for multi-volume data, and these are the cases where you would want slice timing information.

  • The biggest issue I have with current Vida is that there is no versioning, so it is hard to detect hardware changes. All Vida data I have seen reports "XA10A", though it appears there have been some software steppings that are not reported. Every site should log software and hardware changes as possible regressors and as a way to spot performance regressions.

  • Based on a limited sample, assuming a recent XA10A stepping, with enhanced export, without Siemens de-identification tag reduction, the following list are the missing tags relative to an E11C Prisma. Some of these may depend on user setup (e.g. department name), many are public tags where the absence is unexpected (e.g. imagingFrequency; ScanOptions; SAR; SequenceVariant), and some are private tags or CSA ascii fields that are not available in all Siemens products (ConsistencyInfo), others might be missing since I do not have samples (DelayTime, though I think this can be inferred from slice times), several are probably somewhat redundant or poorly defined (Vida does not report PhaseEncodingSteps but it does report EchoTrainLength; DwellTime reported on Prisma may have same unusual interpretation @xiangruili reports for Vida). Regardless, I think this is the list of tags I have never seen from a Vida:

~~Imaging Frequency~~ (substitue: 0018,9098) 
~~interpolation~~ (substitue 0021,1158)
InstitutionalDepartmentName
ScanningSequence
~~SequenceVariant~~ (substitute: 0021,1158)
ScanOptions
SequenceName
SAR (possible substitute: 0018,9179; 0018,9181 for IEC_HEAD)
BaseResolution
ShimSetting
DelayTime
PhaseResolution
PulseSequenceDetails
PhaseEncodingSteps
DwellTime
ConsistencyInfo
WipMemBlock

@captainnova
Copy link
Collaborator

My apologies, the ] is there, I just failed to see it.

     "PhaseEncodingAxis": "j",
     "ImageOrientationPatientDICOM": [
                1,
                0,
                0,
                0,
                1,
                0       ],
      "InPlanePhaseEncodingDirectionDICOM": "COL",

Obviously my eyes had a lot of downward momentum after scanning through the list elements, and I missed it despite double checking with XA10 and Philips. I must have had blinders on.

Rob

@mharms
Copy link
Collaborator

mharms commented Oct 25, 2018

Thanks. Very helpful to have that listing in one place of what is currently "missing".

@BrainStormCenter
Copy link

BrainStormCenter commented Nov 1, 2018 via email

@BrainStormCenter
Copy link

BrainStormCenter commented Nov 1, 2018 via email

@neurolabusc
Copy link
Collaborator

@BrainStormCenter -

  1. Jeff acquired some SMS data, so no need for a new dataset.
  2. The Vida does not populate several useful DICOM fields (e.g. imagingFrequency; ScanOptions; SAR; SequenceVariant). These are pubic fields and not part of the CSA header. They are not required by the DICOM standard, so not a violation. Hopefully in the future Siemens can include these tags.
  3. As long as you save in enhanced format without Siemen's in-built de-identification you should get usable data for DTI (as well as resting state, fMRI, T1, T2).

I am closing this issue, as I think dcm2niix converts all the data Siemens provides us with.

@lucijj
Copy link
Author

lucijj commented Nov 1, 2018

OK, sorry this took so long. Sickness was plentiful and scanner time was scarce. I think I was able to get everything you wanted. a straight, small EPI, and the same scan with PE polarity reversed, with interpolation, with reversed slice order (H-F -> F-H), and with the head turning during the scan. I also got a small 6 direction DTI and another with the PE polarity reversed. These data are publicly shareable, and can be obtained at the following URL:

https://utexas.box.com/shared/static/g8mrr5cgxldepixok9c8agg0vtvsczlo.zip

@xiangruili
Copy link

@neurolabusc
Those 3 fields can be found in SharedFunctionalGroupsSequence.
ImagingFrequency
MRImagingModifierSequence->TransmitterFrequency

SAR
MRTimingAndRelatedParametersSequence->SpecificAbsorptionRateSequence

SequenceVariant (I found a private one)
CSASeriesHeaderInfo -> SequenceVariant : SK

neurolabusc added a commit that referenced this issue Nov 2, 2018
@neurolabusc
XA10A tag substitutes (#240)
5b696cc
@neurolabusc
Copy link
Collaborator

@lucijj and @xiangruili thanks for the sample dataset and suggestions. Latest build now detects substitutes for Interpolation, ImagingFrequency, SequenceVariant. I amended the table above with ~~ to denote changes and substitutions.

@BrainStormCenter
Copy link

BrainStormCenter commented Nov 2, 2018 via email

yarikoptic added a commit to neurodebian/dcm2niix that referenced this issue Dec 3, 2018
@yarikoptic
Merge tag 'v1.0.20181114' into debian
1653ef2 
* tag 'v1.0.20181114': (70 commits)
  New stable release: update documentation (v1.0.20181114)
  Avoid using GDCM's internal OpenJPEG library.
  Discriminate trace from raw GE DWI scans (rordenlab#245)
  Restore Philips enhanced (rordenlab@3c31d18)
  Clean up readme, interpolation errors less verbose, add date-of-birth to BIDS (requires '-ba n')
  Not all GE DWI with b>0 bvec=0 are trace (rordenlab#245)
  calculate Zmm for enhanced DICOM w/o 0018,0088 (rordenlab#241)
  Kludge for Siemens MoCo slice timing interfered with CMRR fix (CMRR-C2P/MB#29)
  Bruker enhanced 4D data (rordenlab#241)
  XA10A tag substitutes (rordenlab#240)
  Huffman tables repeated for RGB planes for jpg_0XC3 (rordenlab#244)
  change MoCo naming and derived detection (rordenlab#243)
  Update dcm_qa_nih submodule.
  Update dcm_qa_nih submodule.
  Update dcm_qa submodule.
  experimental dcm4che enhanced support (rordenlab#241)
  Vida partial Fourier update (rordenlab#240)
  XA10A Vida dwell time (rordenlab#240)
  More XA10A Vida header info (rordenlab#240)
  UIH bvecs (rordenlab#225 (comment))
  ...
@mharms
Copy link
Collaborator

mharms commented May 10, 2022

Regarding #240 (comment):
I know this thread is long closed, but I just wanted to comment for the record that Echo Spacing and Dwell Time are not the same thing (an unfortunate terminology confusion introduced by FSL), but both have value. So I would argue that 0021,1142 ("RealDwellTime") in XA DICOMs should continue to included in the sidecar json's, consistent with the extraction of that value in VE line DICOMs.

neurolabusc added a commit that referenced this issue May 10, 2022
@neurolabusc
Report DwellTime for Siemens XA (#240)
0af5159
@neurolabusc
Copy link
Collaborator

@mharms the development branch now reports 0021,1142 as DwellTime in the BIDS sidecar. I convert the value to seconds. Please validate.

@mharms
Copy link
Collaborator

mharms commented May 11, 2022

Looks good. Thanks for restoring.

yarikoptic added a commit to neurodebian/dcm2niix that referenced this issue Apr 29, 2024
@yarikoptic
Merge tag 'v1.0.20220720' into debian
56f9947 
* tag 'v1.0.20220720': (65 commits)
  GE Direct field mapping (TE1/TE2) (rordenlab#617)
  GE Direct field mapping (TE1/TE2) (rordenlab#617)
  Issue 618 (rordenlab#618)
  Update notes
  Siemens XA30 ASL parameters and ImageTypeText 0021,1175
  Reset PET values for classic DICOMs (rordenlab#616)
  PostLabelDelay for XA30, FrameDuration is only for 4D datasets (rordenlab#616)
  shims are signed (rordenlab#608)
  AcquisitionVoxelSize before any interpolation or resampling within reconstruction or image processing
  Add AcquisitionVoxelSize tag for Siemens ASL (rordenlab#608)
  Store GE ShimSetting as array (rordenlab#608)
  GE sequence details (rordenlab#608)
  Philips slice timing notes
  Verbose scan options (issue 606)
  Change scanOptions
  scan options is long string, fix bvec rejection (rordenlab#606)
  Ignore non-spatial physio data (rordenlab#606)
  Flipping Y also flips sign of determinant
  Better Siemens XA support (rordenlab#606)
  Report DwellTime for Siemens XA (rordenlab#240)
  ...
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
enhancement
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
7 participants
@mharms @lucijj @iamdamion @neurolabusc @captainnova @BrainStormCenter @xiangruili