outline
DESCRIPTION
Outline. Goals Use Cases Ontologies Best Practices and Modeling Issues Collaboration with BioRDF Next Steps: Clinical Observations Interoperability. BIONT Goals. Develop best practices around crucial questions related to creation and use of ontologies: What is an ontology? - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/1.jpg)
Outline• Goals
• Use Cases
• Ontologies
• Best Practices and Modeling Issues
• Collaboration with BioRDF
• Next Steps: – Clinical Observations Interoperability
![Page 2: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/2.jpg)
BIONT Goals• Develop best practices around crucial questions related to
creation and use of ontologies:– What is an ontology?– How should one represent information in an ontology?– Ontology lifecycle: How should ontologies be created, used, accessed,
maintained and evolved?
• Develop use cases spanning the Bench to Bedside Spectrum
• Identify best practices and methodologies in design and development of various ontologies across the Bench to Bedside Spectrum
• Collaboration with various Task Forces: BioRDF, ACPP, SWAN
![Page 3: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/3.jpg)
Use Cases
• Parkinson’s Disease Use Case
• Combined AD – PD Use Case
• Patient Recruitment Use Case
![Page 4: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/4.jpg)
Use Case: Parkinson’s Disease• Description of Parkinson’s Disease and Information Needs from different
perspectives:– Systems Physiology View– Cellular and Molecular Biologist View– Clinical Researcher View– Clinical Guideline Formulator View– Clinical Decision Support Implementer View– Primary Care Clinical View– Neurologist View
• Available at:– http://esw.w3.org/topic/HCLS/ParkinsonUseCase
• Developed by:– Don Doherty– Ken Kawamato
![Page 5: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/5.jpg)
Clinical Research ProtocolEligibility Criteria:
- Inclusion- Exclusion
EMR DATA
Meds Procedures
Diagnoses Demographics
…FailPassPass5/8 criteria met
Yes0033333
…………………
Pass
Pass
Criteria #3
(Pass/Fail/ Researcher Needs to Evaluate)
…
…
…
FailPass3/8 criteria met
No 0022222
Pass
No Criteria #2
(Pass/Fail/ Researcher Needs to Evaluate)
Pass 6/8 criteria met
Yes0011111
Criteria #1
(Pass/Fail/ Researcher Needs to Evaluate)
# Criteria Met / Total Criteria in Protocol
Potentially Eligible for Protocol
Patient MR #
Research Coordinator selects protocol for patient screening:
Research Coordinator views list of patients and selects which ones to approach in person for evaluation and recruitment.
Clinical Evaluation and Recruitment
Research Eligibility Screening Use Case, 9-24-2007
![Page 6: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/6.jpg)
Ontologies
• Biomedical Research– Parkinson’s Disease Ontology
• Clinical Reserch– Problem Oriented Medical Record Ontology– Study Data Tabulation Model
• Clinical Practice– Detailed Clinical Models– ACPP Ontology
![Page 7: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/7.jpg)
Parkinson’s Disease Ontology
![Page 8: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/8.jpg)
POMR Ontology
![Page 9: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/9.jpg)
ACPP Ontology
![Page 10: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/10.jpg)
![Page 11: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/11.jpg)
Collaboration with BioRDF
• Collaboration and contribution to the joint AD – PD Use Case
• BIONT: – Top Down (Domain?) Ontology Construction– Driven by Use Case
• BioRDF: – Bottom up (Data?) driven Ontology construction– Driven by the need to provide a thin layer to
“integrate” the islands
![Page 12: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/12.jpg)
Design Issues in Parkinson’s Ontology
• Using classes vs relationships– UHCL-1 transcribed_into Dardarin
• Using instance-of vs subclasses– UHCL-1 subclass-of Gene vs UHCL-1 instance-of Gene
• Granularity/Specificity of relationships– AllelicVariant causes Disease, vs LRR2KVariant causes Parkinson’s Disease
• Uncertainty– The discovery that genetic mutations in the alpha synuclein gene could cause Parkinson's disease in
families
• Multiple Domains/Ranges– Property: associated_with, Domains: Pathway, Protein, Ranges: Cell, Biomarker
• Default Values– Default function of proteosomal pathway is protein degradation
• Ontology Inclusion and Modularization– NeuroNames, Enzyme Commission, MeSH
• Higher Order Relationships– Association between a Gene and a Disease in the context of a Study
![Page 13: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/13.jpg)
Best Practices for POMR Ontology
• Mappings to foundational ontologies to facilitate ontological commitment
• Adoption of consient terminology for ontological constructs
• Avoidance of constructs which denote cognitive representations (skos:Concept)
• Careful use of partial and complete class axioms • Clear seperation of temporal semantics • Exhaustive disjointeness
![Page 14: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/14.jpg)
Problems and Issues for POMR Ontology
• Expressing periodic time intervals with OWL Time
• No known URI-based naming convention (or OWL export) for SNOMED CT terms
• Lossy semantic transformation from HL7 to RDF
• No feasible means of reasoning over very large ontologies (GALEN, DOLCE, etc..)?
![Page 15: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/15.jpg)
Hard issues
• What level of model becomes a Java class?
• How do you make models easy to use in Java?
• Opposition to this level of detailed models
• Modeling of concepts and quantitative values in a single language/paradigm
• Huge diversity of modeling styles: how to be consistent?
• Defining computable connections between model and externally defined terminology
• Large number of models needed
![Page 16: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/16.jpg)
Next Steps: Clinical Observations Interoperability
• Information Models– DCM– SDTM– BRIDG
• Terminologies– Snomed– NCI Thesaurus– MedDRA
• Re-use and alignment of these models for interoperability
![Page 17: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/17.jpg)
Some tentative proposals
• Two Organizations of Activity
• Technology Driven Organization– BioRDF– BIONT
• Application Driven Organization– ACPP– SWAN– Scientific Publishing
![Page 18: Outline](https://reader036.vdocument.in/reader036/viewer/2022062720/5681340f550346895d9b0163/html5/thumbnails/18.jpg)
Proposed Reorg?• Discovery
– BioRDF– BIONT– URIs– Rules ...
• Development– BioRDF– BIONT– URIs– Rules
• Secondary Uses Of Healthcare Data– BioRDF– BIONT– URIs– Rules …