1 / 8

Creating RDF Representations of Patient Data

Creating RDF Representations of Patient Data. Diabetes Patient Data.

garin
Download Presentation

Creating RDF Representations of Patient Data

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Creating RDF Representations of Patient Data

  2. Diabetes Patient Data 54-year-old man with diabetes was referred to the Hypertension Clinic in Jichi Medical School Hospital with symptoms of dizziness and orthostatic intolerance. In 2002, he started insulin therapy (Hemoglobin A1c 9 %), and in 2004, he was started on a once daily dose of valsartan 80 mg for hypertension. He sometimes complained of dizziness and fainting while standing. On 13 February 2005, he fell down while riding a bicycle. As the clinic BP was 84/60 mm Hg on 14 March, and the valsartan was stopped. He was referred to otolaryngology and neurology, but no significant abnormalities were detected.

  3. URI subject? or has_symptom? age? “Collection of Symptoms” “54 years” _2 _1 “Dizziness” “Orthostatic Intolerance” maps_to maps_to Snomed URI2 Snomed URI1

  4. Design Choices • What are the nodes of this graph? • What are the types of the nodes of this graphs (RDF(S)/OWL Classes)? • What are the labels of the edges of this graph? • How do we decide the classes and instances? • How do we deal with Industry Standards such as RIM/CDA/SDTM/DCM + Snomed/NCI Thesaurus etc.? • Can node types and edge labels be derived from these standards? • Can we have specialized edge (“maps_to”) to link nodes in this graph with Snomed/NCI Thesaurus concepts? • How do we choose these to ensure interoperability across multiple standards? • How do we represent temporal data and information? • How do we capture abstractions such as a clinical event, a patient encounter?

  5. Name Space Issues • Proposal: Create Namespaces for each of the standards • hl7rim: HL7/RIM • dcm: Detailed Clinical Models • cda: Clinical Documentation Architecture • sdtm: Study Data Tabulation Model • snomed: Snomed • ncit: NCI Thesaurus

  6. How do we scope this effort?

  7. Proposed Tasks/Groups • HL7/RIM: Helen, Vipul, Dan • DCM: Tom Oniki, Stan Huff • Realist Approach: Kerstin, Bo, Alan • SDTM: Rachel, Jennifer • POMR: Vipul

  8. Examples of Eligibility Criteria Example Patient Data Various Standards HL7, CDISC, etc. RDF/OWL Graphs

More Related