Menu
▾
▴
Re: [Obi-coord] Common biomedical imaging formats and ontology referencing
|
From: Daniel R. <ru...@me...> - 2007-06-15 02:23:12
|
<html> <body> Fortuitous you should ask...I can tell you about DICOM. DICOM and DICOM-SR allow you to specify coded values in certain DICOM header fields, wherein you specify a "coding scheme designator" (the name of the vocabulary) and the code (the ID for the term in the vocabulary).<br><br> When I last checked, OME didn't explicitly support controlled terminology but were interested in it.l<br><br> IN terms of handling semantic details in images, one very related effort you should follow is the Annotation and Image Markup (AIM) project of caBIG--a project that will define an ontology and standard syntax for describing image metadata and software tools for serializing annotations to DICOM-SR, XML, and HL7-CDA. The goal of this is to capture a rich set of semantic annotations on images that can be incorporated into clinical PACS workstations while interoperating with other health info systems and even Web apps. RadLex will be referenced in radiology images in AIM annotations. There should be something concrete to see on AIM in terms of deliverables by November.<br><br> Regards,<br> Daniel<br><br> ___<br> <x-sigsep><p></x-sigsep> Daniel Rubin, MD, MS<br> Clinical Asst. Professor, Radiology<br> Research Scientist, Stanford Medical Informatics<br> Scientific Director, National Center of Biomedical Ontology<br> MSOB X-215<br> Stanford, CA 94305<br> 650-725-5693<br><br> <br> At 05:24 PM 6/13/2007, William Bug wrote:<br> <blockquote type=cite class=cite cite="">Thanks, Jeff.<br><br> So in regards to the issue on the call re: getting an overview at the OBI meeting of how some of the established export/exchange formats intend to reference investigational ontology entities (MAGE v1.1, FuGE, etc.) - for the bioimaging domain, it would be very helpful to have a very brief overview of how the following image data export/exchange formats deal with the semantic details related to image acquisition and data transformation:<br> 1) DICOM (<a href="http://medical.nema.org">http://medical.nema.org</a>/)<br> 2) NIFTI (<a href="http://nifti.nimh.nih.gov">http://nifti.nimh.nih.gov</a>/)<br> 3) XCEDE (<a href="http://www.nbirn.net/downloads/xcede/index.shtm"> http://www.nbirn.net/downloads/xcede/index.shtm</a>)<br> 4) Open Microscopy Environment [OME] (<a href="http://www.openmicroscopy.org">http://www.openmicroscopy.org</a> /)<br> 5) The Mayo ANALYZE format since its still wide-spread use (<a href="http://www.mayo.edu/bir/Software/Analyze/AnalyzeTechInfo.html"> http://www.mayo.edu/bir/Software/Analyze/AnalyzeTechInfo.html</a>)<br> 6) ANFI Dataset format (<a href="http://afni.nimh.nih.gov/afni/doc/misc/afni_startup"> http://afni.nimh.nih.gov/afni/doc/misc/afni_startup</a>)<br><br> I'm not suggesting we want to take up time reviewing this issue of ontology referencing for all of these formats, but these are the major players, and it would be very helpful to get a sense of how provenance-related (acquisition/processing) semantic details are handled. From my review and work with the formats in the past, my sense is some of the semantics is embedded in the format syntactic structure, and other more granular semantic metadata ends up in structured fields.<br> <br> Daniel Rubin can also provide some insight on how RadLex is expected to be referenced in radiological images (mostly DICOM, I assume), which can provide some insight into how acquisition & processing (data transformation) related info would be referenced. I believe there may also be some relevant work being done within the NAMIC NCBC, as well as on the Neuroimaging Tools and Resource clearinghouse (<a href="http://www.nitrc.org">http://www.nitrc.org</a>/).<br><br> Cheers,<br> Bill<br><br> On Jun 13, 2007, at 12:45 PM, Jeffrey S. Grethe wrote:<br><br> <blockquote type=cite class=cite cite="">Hi Helen,<br> Was on the call this morning, however, wasn't able to unmute (dealing <br> with my daughter) to provide some more information about NIFTI...<br><br> NIFTI (<a href="http://nifti.nimh.nih.gov">http://nifti.nimh.nih.gov</a>/) is the 'Neuroimaging Informatics <br> Technology Initiative' and has mainly focused on data formats/ <br> standards for neuroimaging. The National Institute of Mental Health <br> and the National Institute of Neurological Disorders and Stroke are <br> joint sponsors of this initiative. NIFTI brought together the major <br> tool developers and now most (if not all) of the major tools which <br> are used for neuroimaging analysis (in the research domain, DICOM is <br> the primary in the clinical domain) support NIFTI.<br><br> Cheers,<br> Jeff<br><br> On Jun 13, 2007, at 9:18 AM, Helen Parkinson wrote:<br><br> <blockquote type=cite class=cite cite="">Hi<br><br> most of this is relevant to the dev list. Anyone object to me posting<br> the summary there too?<br><br> cheers<br><br> Helen<br> ---------------------------------------------------------------------- <br> ---------------------------------------------------------------------- <br> ----------------------------<br><br> Summary. OBI coordinator's call 13 June 2007.<br><br> 1. Much discussion on agenda. GF will post new agenda on wiki<br> 2. Branch editors to make it clear that they are<br> 3. Wiki migration hit a few snags. Trish has sent info, report more <br> problems<br> 4. Tina and Trish will work on poster, they invite help from others.<br> 5. GF will look into some video/voice conferencing for the workshop.<br> 6. Need a list of anyone who wants remote access for the upcoming<br> workshop and their preferred tehchnology - skype etc- add to the wiki<br> please or email to GF or Mervi if have probs editing the wiki.<br> 7. Add phone no on the wiki to get a wireless acct during the meeting.<br> No phone number no wireless in NIH.<br><br> OBI coordinator's call 13 June 2007.<br> ------------------------------------<br> Gilberto, Trish, Helen, Daniel, Chris, Jennifer, Bjoern, Liju, Bill<br><br> <br> Agenda items -<br><br> ISMB poster<br> Agenda for workshop.<br> Wiki migration issues<br> Branch leaders<br><br> <br> HP:Subdirs of wiki have gone missing.<br> TW:mostly relevant for biomaterial<br> DS:sent suggestions for order of headers<br> TW:James has made some changes<br> LF;is reln branch a branch, asked to have a space for reln branch<br> TW:moved that on old wiki, is on the new wiki<br> HP:branch page editors will need to do some tidying. All uploaded <br> files<br> are supposed to be there. Need to find out how to link back<br> CS:we can't support making all the changes, each branch needs to make<br> the changes. We will respond to issues. Branch leaders should make <br> changes.<br> BB:The wikipage syntax<br><br> AA:James to post a summary of wiki formatting pages.<br> TW:sent a cheat sheet that I forwarded along<br><br> Workshop:<br> HP has sent out dates for the workshop for data transformation. <br> Open to<br> suggestions about the workshop agenda.<br><br> CS:When we address the agenda for the OBI then we need to schedule <br> some<br> time to discuss this<br><br> HP:branch editor leads could be more transparent.<br> DS:also a problem for reln branch as need to know how to talk to<br><br> CS:Have been keeping tabs on this will mail it out<br><br> TW:Plan hasn't had a discussion about this for e.g.<br><br> CS:ISMB poster - we have one, do we have anyone who is leading the <br> effort,#<br><br> TB:Trish and I are working on that, people can join in.<br> TW:sent out graphic on work flow - handy for the poster, some of <br> the slides<br><br> BB:these are things people want to know about, deliverables, and meta<br> data as well. Examples of two or three different types of experiments<br> fleshed out with OBI - dummied up.<br><br> YW:would be nice to take an example across domains<br><br> BB:thread on lists - issue of how use ontology references in <br> context of<br> exported data e.g. MAGE 2.0 - also see in Birnlex, want to set up <br> syntax<br> and rules of how make use of ontologies and data instances. Have one<br> example, MAGE might be a good one.<br><br> HP:suggest asking Allyson as she's using FUGE and MAGE2 isn't as far<br> along as MAGE2. MAGETAB migt be an interesting case<br><br> BB:Yes. Important to have examples.<br><br> TW:Maybe MAGETAB, in my mind MUGE/MAGETAB doesn't matter if <br> annotated in<br> the same way<br><br> BB:Matter of syntax. Mlchael Miller has been putting out feelers in <br> the<br> context of MAGE2.0 based on FUGE. In BIRN using MAGE1.1 for images and<br> microarray data. Still working through referencing the ontology <br> content.<br><br> CS:Bill, topic is the poster, hearing 2 issues, what should we put on<br> the poster. Made suggestion of a data instance, second issue develop<br> application - good idea for a workshop discussion next month. Lots of<br> suggestions. need to have someone identified to get people to help <br> you.<br><br> TB:Will get to people through the list and get to people next week. <br> Will<br> do in adobe. OK? TW and TB will discuss that offline<br><br> Workshop Agenda.<br><br> GF:Monday am meeting with programme directors. Chris sent out <br> feedback,<br> 10 am start is OK?<br><br> CS:No-one asked about the time. Peter Good will invite people he knows<br> and has ent the poster abstract around to explain.<br><br> GF:Sherri di Coronado will attend<br><br> CS:Also Cheryl Craft, Alison from NIAAD.<br><br> GF:This one will take all Monday am<br><br> From Monday PM possible agenda items:<br><br> 1. Sumary of AA and milestones from SD<br> 2. Update on branch status<br> 3. Discussion of the agenda for rest meeting<br> 4. Cross development issues<br> 5. Maintenance of terminologies by part. communities - within OBI - <br> r or<br> separate<br> 6. Develop agenda for dt workshop in Hinxton<br> 7. Discussion on an application example<br><br> Also allocated time for ontology development.<br><br> CS:That's a lot on Monday pm<br><br> HP:Suggest that the Agenda for the dt workshop can go later<br><br> GF:could move to thurs pm. Application example, same story, Weds pm.<br><br> BB:will need some time to discuss this. If anyone has applications or<br> designs that would be a good forum e.g Allyson<br><br> CS:In terms of optional ppt, we've had new communities. Frank can't <br> make<br> this one, we do have Clinical Ontology - should we get a ppt -we had<br> that last time.<br><br> BB;If there's an update? Links with OBI<br><br> CS:we do have the item of maintenance, might be a good point<br><br> BB:still don;t get if CTO is separate<br><br> <br> BP:We need as much time as poss for the development part. Keep Tuesday<br> to Fridat free.<br><br> TW:OK but we have some outstanding things. No community wide idea of<br> what people will do - contribute to OBI - e.g. PSI, MO etc. Fragmented<br> nature is really difficult<br><br> C:need to be realistic about the workshops, can be useful to take a<br> break within a constrained time period<br><br> GF:Item of maintenance, we can add for some other time. The other item<br> cross development issues could come earlier - affect work for rest of<br> workshop<br><br> TW:Did the reln branch decide when we will add the relns into the <br> file?<br> I am not sure when that was to happen. Could do that before the <br> workshop.<br><br> LF:we are trying to tree the relns from branches and we are doing<br> examples. We will have something preliminary, have 2-3 people only on<br> the calls.<br><br> GF:Should this be a discussion item?<br><br> TW:Or tues/thurs vocab development.<br><br> LF:Agrees T/T development<br><br> LF:will we split into branches?<br><br> GF:I think that's still open? Someone commented that we should work <br> as a<br> group. Don't recall a decision.<br><br> LF:One special occasion, take adv of the whole group. Good to meet <br> together.<br><br> CS:shall we pick a branch and work our way down.<br><br> GF:We should work in a group in a branch that we are not alrady <br> working<br> on sep.<br><br> LF:We should also spend some time on that.<br><br> TB:each branch a ppt - 2 h discussion for each branch? Want to get<br> through all branches. Need to not spend 3 days on one branch that <br> has a<br> lot of problems.<br><br> BP:Want to go through all branches by ppt<br><br> general agreement.<br><br> GF:status from dev branches, we will get into details should we keep<br> them as a quick item or into each as the workshop progresses<br><br> CS:quickly get a status from each, Monday pm might not be the best <br> time.<br><br> TB:15 mins for each branch ppt. 30 mins total for discussion.<br><br> LF:demos will be better in protege<br> TB:some will have excel sheets.<br> BB:as people present this will be dynamics and schedule things <br> later in<br> the workshop.<br><br> TW:are we done with the OBI agenda. What about the ppt on OBI and <br> MIBBI<br><br> GF:was related to programme director presentation.<br><br> LF:do all branches know how to do the merging, Might be useful to <br> have a<br> demo for 30 mins<br><br> GF:can do that in wkshop proper.<br><br> AA:GF will post the revised agenda on the wiki.<br><br> CS:Is video conferencing possible?<br><br> GF:Ryan brought this up, as he can't attend.<br><br> BB:helpful for me, can only do the first 2 days<br><br> GF:we can use centra, need to check on video conferencing<br><br> BB:less important than voice.<br><br> GF:we can try phone conferencing, for outside US need an operator. Can<br> be problematic.<br><br> TW:ppt on MIBBI, FUGE and OBI for the programme dirs. Who's doing <br> them,<br> what's the message, what do we want from the programme dirs?<br><br> CS:The general idea was to take adv of location<br> 1. Awareness raising for granting agencies when developing RFA<br> e.g. electrophysiology ontology - could interface with OBI<br> Make everyone in funders as they think of new directions to think <br> of OBI<br> 2. What kinds of things would like to get funding for - meetings, <br> hiring<br> people etc.<br> We have 2h, select audience. Made sense to have OBI and MIBBI and FUGE<br> in wider stds effort and context.<br> Wearing new hat as president of MGED would also like to cover MGED<br> contribution. Haven't talked to Susanna directly- she'll do MIBBI and<br> FUGE and I can do OBI or someone else can do it, and also will talk<br> about MGED. Any volunteers?<br><br> BP:no that sounds great.<br><br> BB:Sp many communities, at least 2 major imaging efforts in neuro -<br> might be worth Daniel or Jeff presenting? E.g, NIFTI EXCEED etc. No<br> society for MGED - imaging cuts across a lot of disciplines. Also<br> willing to present on OBI<br><br> GF:NIH network people need phone info - need to add it to the wiki?<br><br> TW:Anything in the wiki can be indexed<br> BB:take a screen shot.<br><br> <br><br> ---------------------------------------------------------------------- <br> ---------<br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> <br><br> ---------------------------------------------------------------------- <br> ---<br> This SF.net email is sponsored by DB2 Express<br> Download DB2 Express C - the FREE version of DB2 express and take<br> control of your XML. No limits. Just data. Click to get it now.<br> <a href="http://sourceforge.net/powerbar/db2/"> http://sourceforge.net/powerbar/db2/</a><br> _______________________________________________<br> Obi-coord mailing list<br> <a href="mailto:Obi...@li..."> Obi...@li...</a><br> <a href="https://lists.sourceforge.net/lists/listinfo/obi-coord" eudora="autourl"> https://lists.sourceforge.net/lists/listinfo/obi-coord</a></blockquote> <br> _________________________________________________________________<br><br> Jeffrey S. Grethe, Ph.D. <br> email: <a href="mailto:jg...@nc...">jg...@nc...</a><br> Biomedical Informatics Research Network work: (858) 822-0703<br> Dept. of Neuroscience, School of Medicine fax: (858) 822-0828<br> University of California, San Diego<br> 9500 Gilman Drive, M/C 0715<br> La Jolla, CA 92093-0715<br><br> <br> This e-mail/fax message, including any attachments, is for the sole <br> use of the intended recipient(s) and may contain confidential and <br> privileged information. Any unauthorized review, use, disclosure or <br> distribution is prohibited. If you are not the intended recipient, <br> please contact the sender by reply e-mail/fax and destroy all copies <br> of the original message.<br><br> <br><br> <br> -------------------------------------------------------------------------<br> This SF.net email is sponsored by DB2 Express<br> Download DB2 Express C - the FREE version of DB2 express and take<br> control of your XML. No limits. Just data. Click to get it now.<br> <a href="http://sourceforge.net/powerbar/db2/"> http://sourceforge.net/powerbar/db2/</a><br> _______________________________________________<br> Obi-coord mailing list<br> <a href="mailto:Obi...@li..."> Obi...@li...</a><br> <a href="https://lists.sourceforge.net/lists/listinfo/obi-coord" eudora="autourl"> https://lists.sourceforge.net/lists/listinfo/obi-coord</a></blockquote> <br><br> <br> Bill Bug<br> Senior Research Analyst/Ontological Engineer<br><br> Laboratory for Bioimaging & Anatomical Informatics<br> <a href="http://www.neuroterrain.org/" eudora="autourl"> www.neuroterrain.org</a><br> Department of Neurobiology & Anatomy<br> Drexel University College of Medicine<br> 2900 Queen Lane<br> Philadelphia, PA 19129<br> 215 991 8430 (ph)<br> 610 457 0443 (mobile)<br> 215 843 9367 (fax)<br><br> <br> Please Note: I now have a new email - <a href="mailto:William.Bug@DrexelMed.edu">William.Bug@DrexelMed.edu</a> <br><br> <br><br> </blockquote></body> </html> |