Talk:Semantic Technology Use Cases Catalogue

From PHUSE Wiki
Revision as of 06:50, 5 August 2013 by DanBoisvert (talk | contribs) (Re: Using RDF data in our Workflow -- Mja (talk) 16:53, 25 July 2013 (CDT))
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Proposed Use Cases

  • Representing data standards (e.g. CDISC standards) as an ontology
  • Representing conformance checks in RDF
  • Supporting the development of therapeutic area content development including standards
  • Representing clinical trial data in RDF
  • Representing data exchange standards in RDF
  • Canonical model of Therapeutic Area content represented using SW (RDF, RDFQ, OWL)

Using RDF data in our Workflow -- DanBoisvert 13:28, 18 April 2013 (CDT)

Representing CDISC in RDF is interesting, but how could we go from RDF to SAS (or whatever technology is used to create datasets)

Re: Using RDF data in our Workflow -- Mja (talk) 16:53, 25 July 2013 (CDT)

Here is one way: Use a SPARQL query to obtain a SAS dataset. I have made a SAS macro that simplifies this, and some examples of the use. Please see https://github.com/MarcJAndersen/SAS-SPARQLwrapper - code download https://github.com/MarcJAndersen/SAS-SPARQLwrapper/archive/master.zip. It is an improved version of the SAS code I presented in march.

Re: Re: Using RDF data in our Workflow -- DanBoisvert (talk) 07:50, 5 August 2013 (CDT)

Very interesting. I'll have to try that. Thanks!

Represent the SDRG using Semantic Technology -- DanBoisvert 08:24, 23 April 2013 (CDT)

@kerfors mentioned the possibility of representing the Study Data Reviewer's Guide using semantic technologies. This is beyond me, but sounds interesting. And I quote (from Twitter) "How about a machine process-able Data Guide? In OWL/RDF & SPIN/RDF, see nice SPIN ex. http://topquadrant.com/products/SPIN.html"

Use Semantic Mediawiki -- DanBoisvert 12:23, 25 April 2013 (CDT)

I wonder if there is an opportunity to use the semantic mediawiki extension to be able to produce live (non-production, non-validated) working examples of the use cases we come up with. http://semantic-mediawiki.org/wiki/Help:Using_SPARQL_and_RDF_stores
http://semantic-mediawiki.org/wiki/FAQ


Just a thought.

Metadata Versioning through Semantic Web -- DanBoisvert 11:02, 26 April 2013 (CDT)

From Frederik's presentation, we should write up a good use case for how semantic technologies can be used to version standards. A problem that is faced by all companies.

Constructing and populating semantically rich eCRFs using ISO 11179 registries and the Data Element Exchange profile -- Landenbain 14:37, 7 May 2013 (CDT)

IHE and CDISC have developed an integration profile that allows for the retrieval of metadata from a ISO 11179 metadata repository. This profile can be used to retrieve data from an EHR export document using a metadata enriched eCRF.

Competency Questions

Defining use cases can be a little abstract for people. Another approach (courtesy of Ontology Design 101) is to define a set of questions representing real pain points in your domain that you want this ontology to be able to help you answer. These are referred to as Competency Questions.

If we pose some questions that are currently non-trivial to resolve, then we define something tangible that we are looking to target. Please pose some questions here:

  1. What is the full traceability for this data element in my Analysis Dataset?
  2. What submission datasets depend on the Visit Date in my Schedule of Assessments?
  3. What data elements in my Submission and Analysis Datasets would be affected by my changing the format in this field in my CRF?
  4. What changes were there in the Demography Domain between versions 3-1-2 and 3-1-3 of the SDTM implementation guide?


Re: Competency Questions -- DanBoisvert 09:19, 16 May 2013 (CDT)

5 After working on a product for 10 years and creating "standard" (using the ever changing standards) datasets the whole time how do I integrate the datasets at the end so that I can analyze them all together. What variables, which values, have the same meaning between standards? This sounds like it should be possible now...but as far as I can tell, it's not.
6 Full trace-ability should not stop at analysis dataset, it should go all the way through to table/figure. From a table you should be able to easily traceback and find what variables, which values, which procedures, which statistics were used in producing this number (and what this number really is).

Data Cube -- DanBoisvert 08:29, 14 May 2013 (CDT)

If there is some activity about using RDF data cube/SDMX for clinical trial statistics I would be VERY interested.

Linked Data interface to clinicaltrials.gov -- Kerfors 06:37, 19 June 2013 (CDT) -- Kerfors 06:37, 19 June 2013 (CDT)

We see many nice examples of curated/standardized feeds of CT.gov data, such as http://linkedct.org, http://www.patientslikeme.com/clinical_trials and http://www.clinicalcollections.org/trials/ etc.. Most of them do a good job in turning “strings into things” and a few of them apply the Linked Data principles. However, I don’t think any of them use http-based URIs to identify things such as sponsor organization, clinical sites, clinical investigators, geography, disease, drug, and time.

I argue that we as a community caring for clinical trials data should push back to FDA and NLM to get an official, standardized, linked data interface directly to the CT.gov at source. And yes, also for FDA and NLM to push back to pharma companies to provide standardized data about our trials with URIs to identify things instead of all these text strings. And also if pharma company websites such as http://www.gsk-clinicalstudyregister.com/and http://www.astrazenecaclinicaltrials.com/ did the same.

From the W3C HCLS emailing list: http://lists.w3.org/Archives/Public/public-semweb-lifesci/2013Feb/0067.html