BEP25 - Working solution for spine scans

61 views
Skip to first unread message

Thomas Nichols

unread,
Sep 7, 2020, 10:07:30 AM9/7/20
to Jose Manuel Saborit, Maria de la Iglesia, BIDS Discussion, Daniel Delbarre
Dear Jose & Maria and below-the-neck-BIDS-folk,

Thanks for kicking off work on BEP25, on extending BIDS to other brain regions.  As it happens, I'm involved in a project that also has spine data.  Since there are often multiple spine scans collected at the same time (alongside brain), we've looked to your BEP to see how we could distinguish the body parts.

While I see your BEP proposes a new tag, "bp" for bodypart, we want to work within the existing BIDS framework for now.  Hence we've taken to using the acquisition tag (acq) to label with the body part, with bodypart names drawn (as you suggest) from Body Part Examined DICOM tag, and I assume the table of permitted values (Annex L “Correspondence of Anatomic Region Codes and Body Part Examined Defined Terms”).

It would lead to file names with portions:  "acq-Spine", "acq-CervicalSpine", "acq-ThoracicSpine".  See below for the spine-related portions of Annex L).

I'm writing to solicit input from yourselves and the greater BIDS community about the advisability of this interim solution.  

-Tom


Coding Scheme DesignatorCode ValueCode MeaningBody Part ExaminedSNOMED-RT ID (Retired)BIDS Proposed
SCT122494005Cervical spineCSPINET-11501CervicalSpine
SCT297171002Cervico-thoracic spineCTSPINET-D00F7CervicoThoracicSpine
SCT122496007Lumbar spineLSPINET-11503LumbarSpine
SCT297173004Lumbo-sacral spineLSSPINET-D00F9LumboSacralSpine
SCT39723000Sacroiliac jointSIJOINTT-15680SacroiliacJoint
SCT54735007SacrumSSPINET-11AD0Sacrum
SCT421060004SpineSPINET-D04FFSpine
SCT122495006Thoracic spineTSPINET-11502ThoracicSpine
SCT297172009Thoraco-lumbar spineTLSPINET-D00F8ThoracoLumbarSpine


__________________________________________________________
Thomas Nichols, PhD
Professor of Neuroimaging Statistics
Nuffield Department of Population Health | University of Oxford
Big Data Institute | Li Ka Shing Centre for Health Information and Discovery
Old Road Campus | Headington | Oxford | OX3 7LF | United Kingdom
T: +44 1865 743590 | E: thomas....@bdi.ox.ac.uk
W: http://nisox.org | http://www.bdi.ox.ac.uk

saborit...@externos.gva.es

unread,
Sep 14, 2020, 10:30:39 AM9/14/20
to Thomas Nichols, Maria de la Iglesia, BIDS Discussion, Daniel Delbarre
Dear Tom,

Sorry for answering late and thanks for thinking about giving our
opinion on this adaptation.

We believe that the use of "acq-" like body part tagging is the best
option to work with these images and maintain the BIDS essence. On the
other hand, maybe another way of labeling is not possible in BIDS, so
this method is the only way.

If you need anything, do not hesitate to contact us.

Best regards,

Jose Manuel Saborit Torres.


Thomas Nichols <thomas....@bdi.ox.ac.uk> escribió:

> Dear Jose & Maria and below-the-neck-BIDS-folk,
>
> Thanks for kicking off work on BEP25, on extending BIDS to other brain
> regions. As it happens, I'm involved in a project that also has spine
> data. Since there are often multiple spine scans collected at the same
> time (alongside brain), we've looked to your BEP to see how we could
> distinguish the body parts.
>
> While I see your BEP proposes a new tag, "bp" for bodypart, we want to
> work within the existing BIDS framework for now. Hence we've taken to
> using the acquisition tag (acq) to label with the body part, with bodypart
> names drawn (as you suggest) from Body Part Examined
> <https://dicom.innolitics.com/ciods/raw-data/general-series/00180015> DICOM
> tag, and I assume the table of permitted values (Annex L “Correspondence of
> Anatomic Region Codes and Body Part Examined Defined Terms”
> <http://dicom.nema.org/medical/dicom/current/output/chtml/part16/chapter_L.html#chapter_L>
> ).
>
> It would lead to file names with portions: "acq-Spine",
> "acq-CervicalSpine", "acq-ThoracicSpine". See below for the spine-related
> portions of Annex L).
>
> I'm writing to solicit input from yourselves and the greater BIDS community
> about the advisability of this interim solution.
>
> -Tom
>
>
> Coding Scheme Designator Code Value Code Meaning Body Part Examined SNOMED-RT
> ID (Retired) BIDS Proposed
> SCT 122494005 <http://snomed.info/id/122494005> Cervical spine CSPINE
> T-11501 CervicalSpine
> SCT 297171002 <http://snomed.info/id/297171002> Cervico-thoracic spine
> CTSPINE T-D00F7 CervicoThoracicSpine
> SCT 122496007 <http://snomed.info/id/122496007> Lumbar spine LSPINE T-11503
> LumbarSpine
> SCT 297173004 <http://snomed.info/id/297173004> Lumbo-sacral spine LSSPINE
> T-D00F9 LumboSacralSpine
> SCT 39723000 <http://snomed.info/id/39723000> Sacroiliac joint SIJOINT
> T-15680 SacroiliacJoint
> SCT 54735007 <http://snomed.info/id/54735007> Sacrum SSPINE T-11AD0 Sacrum
> SCT 421060004 <http://snomed.info/id/421060004> Spine SPINE T-D04FF Spine
> SCT 122495006 <http://snomed.info/id/122495006> Thoracic spine TSPINE
> T-11502 ThoracicSpine
> SCT 297172009 <http://snomed.info/id/297172009> Thoraco-lumbar spine TLSPINE

Thomas Nichols

unread,
Sep 14, 2020, 10:50:07 AM9/14/20
to BIDS Discussion, Maria de la Iglesia, Daniel Delbarre
Super! Thanks... we're moving ahead with the "acq" approach for now.  Looking forward to see BEP25 move ahead.

-Tom

--
We are all colleagues working together to shape brain imaging for tomorrow, please be respectful, gracious, and patient with your fellow group members.
---
You received this message because you are subscribed to the Google Groups "bids-discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to bids-discussi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/bids-discussion/20200914163035.Horde.NF08evdQH7lLd-aaBYK_iA4%40webmail.gva.es.

Chris Markiewicz

unread,
Sep 14, 2020, 10:58:47 AM9/14/20
to bids-di...@googlegroups.com, Thomas Nichols, Maria de la Iglesia, Daniel Delbarre
The Microscopy BEP (https://bids.neuroimaging.io/bep031) is also considering how to encode anatomical structures. It looks like they're using "Organ" and "Location" metadata fields, and haven't proposed an entity.

Just a heads up in case there's duplicated effort being made along two fronts.

Chris


Marie-Helene Bourget

unread,
Sep 15, 2020, 11:47:47 AM9/15/20
to bids-discussion

Hi everyone!

Thanks Chris for bringing this discussion to our attention.

In the Microscopy BEP, each sample is identified in the filename with a unique label. Because of that, there is no need for a specific entity in the filename to distinguish between different images of the same subject.

However, this is a question we examined before, and we also thought that people could use the “acq” entity if they wanted to track that information in the filename without it being obligatory (as mentioned in our FAQ: https://docs.google.com/document/d/1Nlu6QVQrbOQtdtcRarsONbX5SrOubXWBvkV37LRRqrc/edit#heading=h.4k5zzopnsbd9).

Best,

Marie-Hélène

Mastaneh Torkamani

unread,
Nov 8, 2024, 7:46:44 AM11/8/24
to bids-discussion
Dear all

I'm re-opening this discussion about inclusion of body parts since development of BEP25 seems to have stopped. 
In a dataset that contains not only in vivo MRI but also ex vivo MRI and multiple microscopy modalities from different brain regions (cortex, hippocampus, or right and left hemispheres), 
what are the recommendations or practices for organizing the raw data while keeping the file levels consistent while keeping different body parts and scans separate? 
For example, 

sub-<label> /
     ses-exvivo /
          bp-<bodypart> /
               anat /
               dwi /
               micr /

or 
sub-<label>

    exvivoMRI (or ses-exvivoMRI?)

     acq-<bodypart> or bp-<bodypart>|

[anat, dwi, etc.]

    micr

acq-<bodypart> or bp-<bodypart>

              [BF, CONF, FLUO] or rather [IHC, nissl, myelin, etc.]?


Thank you in advance for any tips,

Mastaneh 

Oostenveld, R. (Robert)

unread,
Nov 12, 2024, 1:55:02 PM11/12/24
to bids-di...@googlegroups.com
Dear Mastaneh

I suggest you study the microscopy section of the BIDS specification. It prescribes that there are directory levels sub-xx/ses-xx/ followed by modality. So neither bp-xxx nor aq-xxx is permitted as directory level, but acq-xxx can be part of the filename. Furthermore, sample-xxx is required to be part of the filename (for microscopy, not for anatomical MRI).

The BodyPart and BodyPartDetails are coded in the json file (see https://bids-specification.readthedocs.io/en/stable/modality-specific-files/microscopy.html#sample), not in the file name nor in the directory name.

best regards,
Robert



Reply all
Reply to author
Forward
0 new messages