Ontology-Driven Clinical Intelligence A Path from the Biobank to Cross-Disease Research Bruce Pharr | Vice President, Bioinformatics Systems Molecular Medicine Tri-Conference | February 11, 2014 1
May 07, 2015
Ontology-Driven Clinical Intelligence A Path from the Biobank to Cross-Disease Research
Bruce Pharr | Vice President, Bioinformatics Systems
Molecular Medicine Tri-Conference | February 11, 2014
1
Data Barriers to Clinical Research Critical Data is Dispersed in Separate Systems
Considering the vast stores of clinical data available to potential investigators, the actual amount of clinical research performed has been quite modest. At many medical centers, the data are dispersed in separate systems that have evolved independently of one another.
Source: Obstacles and Approaches to Clinical Database Research: Experience at the University of California, San Francisco
Disease A Disease B
Removing the Data Barriers Structured Digital Data with Standardized Metadata and Ontology
Source: Anne E. Thessen and David J. Patterson, Data issues in life sciences, PMC (NIH/NLM) (November 28, 2011).
Disease A Disease B
The discovery of scientific insights through effective management and reuse of data requires several conditions to be optimized:
• Data need to be digital; • Data need to be structured; • Data need to be standardized in terms of metadata and ontology.
Ontology-Driven Clinical Intelligence Structured Data with Standardized Metadata and Ontology
Mosaic™ Ontology-Based Platform
Pre-analytical Data Analytical Data
Lab Test & Analysis
Disease Registry
New Patient
Legacy Disease Database
Legacy Data
Biobank
Patient Data
Patient Data
Remedy Informatics Mosaic™ Platform
Ontology-Driven Clinical Intelligence Remedy Informatics Architecture
Mosaic Engine Functional Layers: Physical, Data Model, Information Model, Ontology, Representation Model
Mosaic Builder Applications Content and Registry Development
Remedy Bioinformatics Biobank Management Informatics
RemedyAMH™ Aggregate, Map & Harmonize
Patient Data
Legacy Data
Patient Data
Disease Registry
New Data Patient
Data
Remedy Informatics
Next-Gen Biobank A Path from the Biobank to Cross-Disease Research
Remedy Bioinformatics Biobank Management Informatics
New Data Patient
Data
Biobank Growth and Upgrade Cycle Drivers for Next-Gen Biobanks
Growth 33% of all biobanks have been installed since the early 2000s (HGP)
• Increase in population genetics studies • Personalized medicine • Genetic information in food safety, forensics and disease surveillance
Upgrade The Cancer Genome Atlas (TCGA) project (2006-8) exposed deficiencies
• Many biobank managers didn’t know exactly what was in their freezers • Some specimens were unfit for analysis • Others had been obtained from patients without adequate consent • The rate of unacceptable shipments from some institutions was 99%
Source: The Future of Biobanking, Laboratory Focus, January 2013
Next-Gen Biobank Management Best Practices Model Mapped to Applicable Global Standards
Patient
Biobank Manage all information about: 1. Specimens, 2. Patients, and 3. Operations throughout:
• Collection • Processing • Storage and Inventory • Distribution
Best Practices Biobank Management Informatics Requirements
• Metadata • Entity Types • Sample Acquisition • Sample and Data Management • Sample Retention and Distribution • Support of Laboratory Processes • User Management • Search • Presentation of Entities • Printing • Reports and Audits • Non-functional Requirements • External Interface Requirements
Best Practices Applicable International Standards and Guidelines
ISBER International Society for Biological and Environmental Repositories. Best Practices for Repositories: Collection, Storage, Retrieval, and Distribution of Biological Materials for Research.
NCI National Cancer Institute. First-generation guidelines for NCI-supported Biorepositories.
BAP Biorepository Accreditation Program (BAP) Checklist – College of American Pathologists (CAP)
21 CFR Part 11 US FDA – Guidelines on electronic records and electronic signatures.
45 CFR § 164.514 US HHS – Other requirements relating to uses and disclosures of protected health information.
ISO 15189 Medical laboratories – Particular requirements for quality and competence.
ISO 17025 General requirements for the competence of testing and calibration laboratories.
MoReq2 European Commission. Model Requirements for the management of electronic records.
OECD Best Practice Guidelines for biological resource centres.
Rec(2006)4 Council of Europe, Committee of Ministers. Recommendation of the Committee of Ministers to member states on research on biological materials of human origin.
Remedy Informatics Mosaic Platform
Mosaic Ontology Purpose-Specific Structured Data Model
Mosaic Engine Functional Layers: Physical, Data Model, Information Model, Ontology, Representation Model
Mosaic Builder Applications Content and Registry Development
RemedyAMH™ Aggregate, Map & Harmonize
Patient Data
Legacy Data
Patient Data
Disease Registry
1. Predefined, Standardized Terminology 2. Domain-Specific Mapped Relationships 3. Permissible Values and Validation Rules
Mosaic Ontology Predefined, Standardized Terminology
Lab Result LOINC
Subject
Units
High End of Normal
Low End of Normal
Confidentiality
Validation Status
Validator
Supplier of Data
LOINC Medical Laboratory and Clinical Observations
Mosaic Ontology Predefined, Standardized Terminology
Disorder SNOMED CT
Assertion
Subject
Severity
Stage
Response to Treatment
Active State
Onset Date
Resolved State
First Diagnosed Date
Confidentiality
Source
Date of Entry
Validation Status
Validator
Supplier of Data
LOINC Medical Laboratory and Clinical Observations
SNOMED CT Clinical Codes, Terms, Synonyms and Definitions
Mosaic Ontology Predefined, Standardized Terminology
LOINC Medical Laboratory and Clinical Observations
SNOMED CT Clinical Codes, Terms, Synonyms and Definitions
ICD Disease Classifications
Gene Ontology Gene Product Characteristics and Annotation
RxNorm Clinical Drug Classifications
CDISC Clinical Protocol, Analysis and Reporting
Has Result
Response to Tx
Evidence for
Cause
Mosaic Ontology Domain-Specific Mapped Relationships
Lab Result LOINC
Subject
Units
High End of Normal
Low End of Normal
Confidentiality
Validation Status
Validator
Supplier of Data
Disorder SNOMED
Assertion
Subject
Severity
Stage
Response to Treatment
Active State
Onset Date
Resolved State
First Diagnosed Date
Confidentiality
Source
Date of Entry
Validation Status
Validator
Supplier of Data
Procedure SNOMED
Subject
Operator
Facility
Start-Stop Time
Urgency Status
Intent
Confidentiality
Source
Date of Entry
Validation Status
Validator
Supplier of Data
Indication
Contraindication
Mild
Moderate
Severe Screening
Diagnostic
Prevention
Therapeutic
Palliation
End-of-Life
Mosaic Ontology Permissible Value and Validation Rules
Disorder SNOMED
Assertion
Subject
Severity
Stage
Response to Treatment
Active State
Onset Date
Resolved State
First Diagnosed Date
Confidentiality
Source
Date of Entry
Validation Status
Validator
Supplier of Data
Procedure SNOMED
Subject
Operator
Facility
Start-Stop Time
Urgency Status
Intent
Confidentiality
Source
Date of Entry
Validation Status
Validator
Supplier of Data
Mosaic Ontology Standardized, Extensible Disease Registry Implementation
Ontology-Driven Clinical Intelligence Cross-Disease Research
• Founded in 2003, privately held. • U.S. headquarters in Salt Lake City, Utah. Development offices in
Menlo Park, California.
• Satellite offices in London, England; Sao Paulo, Brazil; and Munich, Germany.
• More than 120 employees.
• Strategic partnerships with Merck and IMS.
• Developed proprietary Mosaic Platform, an ontology-driven clinical intelligence system scalable to any size enterprise.
• Delivered more than 120 registries to wide range of leading life sciences research and healthcare delivery organizations.
Remedy Informatics
Thanks! – Questions?
Bruce Pharr Vice President, Bioinformatics Systems [email protected] Remedy Informatics www.remedyinformatics.com
Booth 406