-
Notifications
You must be signed in to change notification settings - Fork 165
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.
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
[ENH] Bep 005: Arterial Spin Labeling #652
Conversation
Merging 1.5.0dev changes
single and multi timepoint PASL, PCASL and CASL
Storing should always be as a separate 4D nii file in original acquisition order 2 ancillary files: asl.json and aslcontext.tsv
Often additional scaling should be performed. In BEP005 we define that this should be done during conversion of dicom to BIDS. Thus no scaling information should be left in BIDS (json, nii header,...) to avoid confusion
as often multi timepoint asl is acquired, some fields allow the use of a vector. Vector length should always be the same as aslcontext.tsv number of volumes.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
we should move this PR from patsycle:BEP-005
to bids-standard:bep-005
, so that the ReadTheDocs rendering will work at this address: https://bids-specification.readthedocs.io/en/bep-005/
unfortunately I do not now how we can move the "source" of the changes (patsycle repo to bids-standard repo) ... in the GitHub user interface, only changing the target of the changes (currently bids-standard:master
) seems to be possible.
If there is no way to change the "source" of changes in this present PR, we'll have to open a new one, and will lose all comments from here.
We should decide about that relatively soon, before more comments come in.
EDIT: Other opinions?
cc @bids-standard/maintainers
| **Key name** | **Requirement level** | **Data type** | **Description** | | ||
|-----------------------------------|---------------------------------------------------------------------------------------------------------------------------------|-----------------------------------------|-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------| | ||
| LabelingType | REQUIRED | [string][] | `CASL` , `PCASL` , `PASL`. | | ||
| PostLabelingDelay | REQUIRED | [number][] or [array][] of [numbers][] | This is the postlabeling delay (PLD) time, in seconds, after the the end of the labeling (for `(P)CASL`) or middle of the labeling pulse (for `PASL`) until the middle of the excitation pulse applied to the imaging slab (for 3D acquisition) or first slice (for 2D acquisition). Can be a number (for a single-PLD time series) or an array of numbers (for multi-PLD and Look-Locker). In the latter case, the array of numbers contains the PLD of each volume, namely each `control` and `label`, in the acquisition order. Any image within the time-series without a PLD, for example an `m0scan`, is indicated by a zero. Based on DICOM Tags 0018,9079 `Inversion Times` and 0018,0082 `InversionTime`. | |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
There is a typo "after the THE end of the labeling..."
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Unless I missed something I think that there are some redundancies in some of the new tables with other sections that could be avoided.
src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md
Outdated
Show resolved
Hide resolved
…ata.md Co-authored-by: Remi Gau <remi_gau@hotmail.com>
@patsycle @HenkMutsaerts I pulled the most recent version of the bids-specification into this Pull Request, so all "merge conflicts" should be resolved now. IMPORTANT --> Perform In the process, I removed two paragraphs that you added, because while you were working on them, somebody else refactored that part of the specification, which left us in this situation:
I hope that makes the "merge conflict" situation clear. These are the two parts that I removed that have now to be re-added: part 1
part 2
Once you have re-added these parts ...... we would like to squash and merge all commits into this branch: https://github.com/bids-standard/bids-specification/tree/bep-005 @HenkMutsaerts and @patsycle will then get "write" rights on the bids-specification repository and you can continue on a new PR from that new bids-specification-owned branch to bids-specification-owned "master". I have hinted at that here: #652 Please ask any questions you might have about this git / GitHub process 🙂 |
To be clear, the variability in TE/FA (necessitating arrays instead of single values) is because the possibility that M0/CBF/deltaM scans are in the |
@sappelhoff Sorry, accidently closed PR. But I re-opended! |
@effigies No these can vary in general, in case of advanced ASL sequences such as QUASAR. |
Please continue commenting on #669 |
| M0 | REQUIRED | [number][] or [string][] or [boolean][] | Describes the available M0 information or the location of an `m0scan` file, as either: *Numerical*: a single numerical whole-brain M0 value, for example obtained from CSF in another sequence; *SeparateFile*: `*_m0scan.nii[.gz]`; *Within ASL time-series*: *Boolean: `true`* referring to an `m0scan` within the time-series (as defined within the `*_aslcontext.tsv`); *`false`* referring to the use of the (average) `control` image for M0 calibration. | | ||
| MagneticFieldStrength | REQUIRED | [number][] | Nominal field strength of MR magnet in Tesla. Corresponds to DICOM Tag 0018,0087 `Magnetic Field Strength`. | | ||
| PulseSequenceType | REQUIRED | [string][] | General description of the pulse sequence used for the scan. Allowed values are any combination of 2D or 3D acquisition, with the read-out type, separated by an underscore: for example “2D_EPI”, “3D_GRASE”, “3D_spiral”. | | ||
| EchoTime | REQUIRED | [number][] or [array][] of [numbers][] | The echo time for the acquisition, specified in seconds. Specify either one number for the total time-series, or provide an array of numbers with different values of each volume. Corresponds to DICOM Tag 0018, 0081 `Echo Time` (please note that the DICOM term is in milliseconds not seconds). | |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@effigies Hi Chris, last friday we discussed the option to allow 'array of numbers', with the same length as number of volumes, for EchoTime and FlipAngles. I was wondering how we do this practically? I can easily propose it in the common fields in bep005, is that okay? For validator, should I ask Marco to do the same?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Yes, let's change it in Timing Parameters and RF & Contrast.
I would suggest the text:
The echo time (TE) for the acquisition, specified in seconds. A single value applies to all volumes, or an array provides a value for each volume. Corresponds to DICOM Tag 0018, 0081
Echo Time
(please note that the DICOM term is in milliseconds not seconds). 4D files with variable TE should only be used when volume timing is critical for interpretation of the data, as in ASL or variable TE fMRI sequences.
and
Flip angle (FA) for the acquisition, specified in degrees. A single value applies to all volumes, or an array provides a value for each volume. Corresponds to: DICOM Tag 0018, 1314
Flip Angle
. 4D files with variable FA should only be used when volume timing is critical for interpretation of the data, as in ASL or variable FA fMRI sequences.
Then in the ASL section, we mark them as REQUIRED for ASL sequences.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
@effigies Thanks for the suggested texts. I will add those today.
For Requirement level, I did change it now for MagneticFieldStrenght into 'RECOMENDED, but REQUIRED for Arterial Spin Labeling', and the same for the new MRAcquisitionType field, in imitation of what was done already for EchoTime and SliceTiming. I can do the same for EchoTime and FlipAngle, but I agree that, if more BEP's do this, the requirement level fields might get long. What is prefered, so I can keep it consistent?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think leaving "REQUIRED for Arterial Spin Labeling" is fine. We can reorganize in the future if things start getting hard to read.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I agree! I'll ask Marco to take a look into the validator as well for FlipAngle and EchoTime.
Co-authored-by: Patricia Clement <41481345+patsycle@users.noreply.github.com> Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org> Co-authored-by: Chris Markiewicz <effigies@gmail.com> Co-authored-by: Julia Guiomar Niso Galán <guiomar.niso@ctb.upm.es> Co-authored-by: Remi Gau <remi_gau@hotmail.com>
Co-authored-by: Patricia Clement <41481345+patsycle@users.noreply.github.com> Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org> Co-authored-by: Chris Markiewicz <effigies@gmail.com> Co-authored-by: Julia Guiomar Niso Galán <guiomar.niso@ctb.upm.es> Co-authored-by: Remi Gau <remi_gau@hotmail.com>
* [ENH] Bep 005: Arterial Spin Labeling (#652) Co-authored-by: Patricia Clement <41481345+patsycle@users.noreply.github.com> Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org> Co-authored-by: Chris Markiewicz <effigies@gmail.com> Co-authored-by: Julia Guiomar Niso Galán <guiomar.niso@ctb.upm.es> Co-authored-by: Remi Gau <remi_gau@hotmail.com> * Update table errors * correction for table issues 'common metadata for PCASL/PASL' * correction latin phrases * Corrections tables * correction tables Cases * added link pepolar * deleted 'don't discriminate between types of labeling based on comment Thijs Van Osch * Moved MagneticFieldStrenght asl requirement to common * removed scaling factor info based on comment gllmflndn Removed: all ancillary scaling factors should be taken into account in the conversion to BIDS, which is why BIDS does not provide a separate scaling factor field other than the NIfTI header. * added MRAcquisitionType added in the common - sequence specifics table Reason: for ASL, defining 2D and 3D is required. This used to be added in the PulseSequenceType field, but this should stay a free text field. Therefore, the MRAcquisitionType field is added (based on DicomTag). * removed PulseSequenceType from the ASL part The 2D/3D information needed for ASL, is now moved to the new MRAcquisitionType field. * added required common fields sentence Some Common fields are required for ASL. To stress this out, we added a sentence for the *_as.json and the *_m0scan.json. * adaptation requirements for m0scan.json * adapted requirement level common fields for m0scan * update sentence required common fields for asl.json * Update common field EchoTime * Removed EchoTime from ASL-table * update FlipAngle in common RF and Contrasts * remove FlipAngle from ASL fields * update Dependency table ASL for MRAcquisitionType * update SliceTiming in common fields * remove SliceTiming from ASL tables * added (FA) for FlipAngle * update common RF and Contrast for table check * LabelingPulseDuration and LabelingPulseInterval in ms instead of s * required common metadatafields update RepetitionTime or VolumeTiming to RepetitionTimePreparation * update flipangle for timeseries versus file collections * update echotime for timeseries * update timing table (fence) * update definition BackgroundSuppressionPulseTime * removed common metadata fields from asl table MagneticFieldStrenght, EchoTime, SliceTiming, VolumeTiming, RepetitionTime, FlipAngle, PulseSequenceType * update scaling of asl and m0 files - update scaling description for nifti files - created separate scaling section (to highlight this important quantification factor), and M0 section - moved scaling section and M0 section after aslcontext section * update broken link in EchoTime (Timing Parameters) * update broken link (2) EchoTime * Update broken links FlipAngle * LabelingType changed to ArterialSpinLabelingType - changed in fields - Changed in dependency table - Dependency table: additionally removed repetitiontime/volumetiming parts * Update dependency table based on new M0Type and M0Estimate fields * Update new M0 field strategy M0 field becomes M0Type, with addition field M0Estimate. Also corrected IntendedFor description in m0 specific fields * TEMPORARY: disable 'strict' docs build this allows the build to proceed despite broken links * changed fail_on_warning to true Intermediate solution to get BEP005 build untill the missing link is available * see previous commit * update bep005 with latest master + resolve conflict (#689) * [DOC] Auto-generate changelog entry for PR #677 * [ENH] BEP001 - New entities: inv & mt (#681) * add entity inv * add mt: * add MT specific metadata * Generate entity.md * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org> * Update src/schema/entities.yaml Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org> * table fix by @sappelhoff * Conditional requirement Co-authored-by: Stefan Appelhoff <stefan.appelhoff@mailbox.org> * [ENH] BEP001 - Entity-linked file collections (#688) * Squashed commit of the following: commit 8e1ac5f Author: Stefan Appelhoff <stefan.appelhoff@mailbox.org> Date: Fri Nov 27 11:09:24 2020 +0100 fix links and latin commit 27c9b41 Author: Stefan Appelhoff <stefan.appelhoff@mailbox.org> Date: Fri Nov 27 10:03:51 2020 +0100 remove latin Co-authored-by: Chris Markiewicz <effigies@gmail.com> commit 7d87070 Author: Agah <agahkarakuzu@gmail.com> Date: Mon Nov 23 08:46:47 2020 -0500 Update src/99-appendices/10-file-collections.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> commit b137a9e Author: Agah <agahkarakuzu@gmail.com> Date: Mon Nov 23 08:45:19 2020 -0500 Update src/02-common-principles.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> commit d7897b8 Author: Agah <agahkarakuzu@gmail.com> Date: Mon Nov 23 08:44:40 2020 -0500 Update src/02-common-principles.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> commit 09e9d32 Author: Agah <agahkarakuzu@gmail.com> Date: Mon Nov 23 08:44:15 2020 -0500 Commit suggestion Co-authored-by: Chris Markiewicz <effigies@gmail.com> commit 009a9ab Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Mon Nov 23 08:43:00 2020 -0500 Add file-collecitons appendix to the TOC commit 8f6c4df Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Mon Nov 23 08:41:13 2020 -0500 Address suggestion by @tsalo commit 45210a9 Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Wed Nov 11 14:32:20 2020 -0500 Wording commit 66ba5a5 Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Wed Nov 11 14:31:40 2020 -0500 RECOMMENDED --> MUST for adding an application def to the appdx commit 5611fac Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Wed Nov 11 14:29:21 2020 -0500 Improve the appendix commit 6d3bbea Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Mon Nov 2 16:12:04 2020 -0500 Address suggestions by @tsalo commit bcb0223 Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Tue Oct 27 21:37:26 2020 -0400 [ADD] Link to the appendix commit 44e760f Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Tue Oct 27 21:28:38 2020 -0400 [ADD] Appendix - File collections - For parametrically linked file collections commit 523b1c1 Author: Agah Karakuzu <agahkarakuzu@gmail.com> Date: Tue Oct 27 21:27:52 2020 -0400 [ADD] Parametrically linked file collections - Description * ENH: Link to all BEP-001 entities * STY: Escape asterisk, adjust table widths * Add suggestions by @sappelhoff * Update src/02-common-principles.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> * Update src/02-common-principles.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> * Update src/02-common-principles.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> * Update src/99-appendices/10-file-collections.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> * Update src/02-common-principles.md Thank you @effigies! Co-authored-by: Chris Markiewicz <effigies@gmail.com> * Update src/02-common-principles.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> Co-authored-by: Agah Karakuzu <agahkarakuzu@gmail.com> * [ENH] BEP001 - RepetitionTimeExcitation and RepetitionTimePreparation (#671) * [ENH] Improve TR definitions - Include RepetitionTimePreparation - Include RepetitionTimeExcitation - Add explanation on the former RepetitionTime field. * Typo * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> * ENH: Allow for variable RTE/P * STY: Update table widths * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md * Change ref to DOI @Remi-Gau * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <effigies@gmail.com> Co-authored-by: Chris Markiewicz <effigies@gmail.com> Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * Update config.yml * Update readthedocs.yml Co-authored-by: bids-maintenance <bids.maintenance@gmail.com> Co-authored-by: Agah <agahkarakuzu@gmail.com> Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> Co-authored-by: Chris Markiewicz <effigies@gmail.com> * update required common fields for m0scan.json added EchoTime and FlipAngle in case LookLocker is true * update dependency table if PCASL : LabelingDuration is required if PCASL: LabelingDuration should not be filled in * update depedency table removed 'PASL' LAbelingDuration should not be filled in * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * update description aslcontext Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * update PCASLType and CASLType Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * updated required into REQUIRED for asl.json and m0scan.json (common fields * Added SummaryImages Added images: asl_pasl_boluscutoff_false.png asl_pasl_boluscutoff_true_q2tips.png asl_pasl_boluscutoff_true_quipssII.png asl_pcasl_labeling_pulses.png asl_pcasl_sequence.png * create Appendix 11 for ASL * Added asl_flowchart.png * added subtitles for linking from main specification * Added appendix link for control and label * added appendix for asl sequences in general * added appendix for (P)CASL * added appendix for PASL * added appendix for dependency table (flowcharts) * MNT: Add ASL appendix to TOC * TYPO: correct ASL appendix entry in tocwq * fix md style * Update src/04-modality-specific-files/01-magnetic-resonance-imaging-data.md Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> * update table RF and Contrast * update link Appendix XI to Appendix XI - ASL * corrected link to appendix for PASL * Update aslcontext cases Case 1: changed asl.json to asl.nii[.gz`] Removed: The optional deltam or cbf volumes should be stored and specified as derivative. Case 2: Removed: The optional cbf volumes should be stored and specified as derivative. * Added aslcontext.tsv cases this is moved from main spec bep005 to appendix XI - ASL * Update aslcontext cases - moved three cases to appendix XI - ASL - added link to appendix XI - added: Note that the raw images, including the m0scan, may also be used for quality control. * corrected subtitles cases * improve linebreaks in ASL appendix * fix VascularCrushingVENC link Based on comment of Stefan Appelhof, the Data Type for VascularChrushingVENC was fixed: [number][] or \[array][] of [numbers][] to [number][] or [array][] of [numbers][] * added single column requirement for aslcontext.tsv based on comment of Stefan Appelhof, the need for a single column was added in the introductory text for aslcontext.tsv, before the 3 cases. * Changed VascularCrushingVENC to VascularCrushingVenc * [SCHEMA] Add ASL to schema (#703) * Add ASL to schema files and regenerate entity table. * Fix style issue. * Add fmap-format m0scan to schema. * update M0Estimate fields Added 'Referring to the M0 of blood' * update LabelingDuration Added additional information for LabelingDuration * update labeling.jpg into *_asllabeling.jpg * Change labeling to asllabeling. (#709) * RF: Renumber ASL appendix to XII Co-authored-by: Henk Mutsaerts <henkjanmutsaerts@gmail.com> Co-authored-by: Patricia Clement <41481345+patsycle@users.noreply.github.com> Co-authored-by: Chris Markiewicz <effigies@gmail.com> Co-authored-by: Julia Guiomar Niso Galán <guiomar.niso@ctb.upm.es> Co-authored-by: Remi Gau <remi_gau@hotmail.com> Co-authored-by: bids-maintenance <bids.maintenance@gmail.com> Co-authored-by: Agah <agahkarakuzu@gmail.com> Co-authored-by: Chris Markiewicz <markiewicz@stanford.edu> Co-authored-by: Taylor Salo <tsalo006@fiu.edu>
These are the pre-existing BIDS keys for which ASL has additional requirements:
EchoTime
FlipAngle
MagneticFieldStrength
PulseSequenceType
RepetitionTime
SliceTiming
VolumeTiming
Please continue commenting on #669