14:59:42 RRSAgent has joined #hcls 14:59:46 logging to https://www.w3.org/2024/03/14-hcls-irc 14:59:52 rrsagent, make logs public 14:59:58 Meeting: FHIR RDF 14:59:58 Chair: David Booth 15:04:53 Topic: DICOM 15:05:18 erich: Josh and I attended FHIR imaging group. Read through their docs. They're working on a mapping to FHIR. 15:05:45 ... Done quite a bit. Important, but not the same as castingDICOm to RDF. There's overlap between the models. Need to map them. 15:05:55 ... Still need an official casting of DICOM to RDF. 15:07:03 .. But I was able to convert DICOM to JSON. DCM dump to JSON is based on the spec. Everything is multivalued. 15:07:36 .. Taking that, going into JSON-LD playground, and adding an @context, it turns into RDF as quads. 15:08:19 ... But there's a problem. In the doc, when you put nulls into an array in json-ld they get filtered out, but DICOM requires them to be kept. 15:08:48 ... Null resets values instead of not asserting anything. 15:09:07 ... Want to hold a position in the list. 15:10:20 ... The default vocab covers everything. I got an @context for the spec. But custom tags would not be passed through. But @vocab can give a default. 15:11:06 ... Also considered a differeent approach, 20B triples had perf issues. Need date strings to be dates, etc. 15:11:37 ... But once you have it in there, there isn't much need for the value rep, w extra blank node etc. 15:12:01 ... Another possibility would be to make custom datatypes for DICOM, and tag that onto the triple. 15:12:23 ... Jena and Virtuoso both can map custom datatypes to native types. 15:12:54 ... File comes out looking just like the original DICOM. 15:13:17 ericp: Checked w Gregg Kellogg? 15:14:18 ... In this context null has meaning. 15:14:40 erich: Also, stay w these OIDs vs convert to URN? 15:15:48 eric: Pref the idea of deep curation. FHIR had datatypes that are union types, eg date datetime, and we separatted them to different types. 15:16:08 .. Suspect a similar rationalization would make it more RDF friendly. 15:17:52 erich: SPARQL UPDATE query could transform the versoin w custom datatypes to the form that has properties. 15:18:16 .. Bnefit of matching DICOM form, is that tools could convert it back ti DICOM binary. 15:18:57 ... DICOM was adopted because of ease of using it. DICOM to JSON shows how to map them. 15:19:20 ... If we can transform back it would be good. 15:19:39 detlef: Checked how they transform DICOM sequences? 15:20:29 erich: They have nested array. 15:20:42 ... In DICOM JSON, everything is multivalued. 15:21:15 ... Turns into an RDF list. Played w RDF sequence w explicit index. 15:21:28 ... But JSON-LD only maps to an RDF list. 15:22:12 ... Also for everything w a cardinality of one, everything in DICOM JSON is an array. 15:22:28 ... I've only ever seen this in Drupal. 15:22:55 erip: FHIR itself requires preserving list order. 15:24:47 erich: They say "if a property exists in the source but no value, they want it preserved in the JSON, but no value". That works fine for RDF, except if there a vield w only one value. 15:25:26 ... How to represent that in RDF if there's no value? 15:25:49 detlef: In RDF there are mandatory properties, even if there is no value. 15:27:08 dbooth: Could have a distinquished value meaning null. 15:27:32 erich: Could also reach out to DICOM group about how to move it forward. But could do community group first. 15:31:18 erich: JSON-LD framing is pretty powerful. Mibht convert optimized RDF to JSON. 15:33:04 ericp: One potential choice is to try for something close to what you want, and show exciting use cases. 15:33:38 ... Nice if we can build an ecosystem where all the things in healthcare are RDFy. Tempted to try it and back off if needed. 15:34:38 ericp: Our preprocessing step started to get too complex, not worth it. 15:35:06 ... Re framing, anyplace w nested structures won't work well. They'll come out arbitrarily. 15:35:56 erich: Since JSON-LD exists, if they matched it would be a neat way to go. Would like to try as RDFy as possible. Shold be fully transformable back into the original. 15:36:07 ... Play w Framing to see if it can do it. 15:37:17 erich: Start a W3C DICOM community group? 15:37:45 ericp: You need 3-4 people to say they're interested. 15:38:08 detlef: I'm interested. At the moment our code is proprietary, need to get permission. 15:38:24 erich: I'm happy to publish my stuff. 15:39:06 ... Working to go directly from DICOM to RDF. That's harvesting our covid data. 15:40:10 ... 350TB. 15:40:34 ACTION: Erich to initiate DICOM community group 15:40:49 erich: Makes sense to coordinate w FHIR imaging group. 15:41:07 ... I'll keep sitting in on mtgs. 16:28:44 Present: EricP, Erich Bremer, David Booth, Detlef Grittner 16:28:48 ADJOURNED 16:28:56 rrsagent, draft minutes 16:28:57 I have made the request to generate https://www.w3.org/2024/03/14-hcls-minutes.html dbooth