Theodoros. N. Arvanitis, RT, DPhil, CEng, MIET

TRANSFoRm
Theodoros. N. Arvanitis, RT, DPhil, CEng, MIET, MIEEE, AMIA, FRSM
Biomedical Informatics, Signals & Systems Research Laboratory
School of Electronic, Electrical & Computer Engineering
College of Engineering and Physical Sciences
University of Birmingham
Birmingham Children’s Hospital NHS Foundation Trust
TRANSFoRm is partially funded by the European Commission - DG INFSO (FP7 247787)
1
Knowledge in healthcare
Specific
research
knowledge
•
•
•
•
EHR systems
Wide coverage
Vast quantity
May lack in detail
and quality
Routinely
collected
knowledge
• Clinical trials
• Controlled populations
• Well-defined questions
Actionable
knowledge
• Distilled scientific
findings
• Usable in clinical
practice
• Decision support
2
The challenge of representing knowledge in an
interoperable computable form
• Developing a user understandable, computable and extensible
knowledge representation scheme for capturing clinical trials’
concepts and information (knowledge)
– with a multilingual support
• The foundation of interoperability lies with a shared understanding
of concepts and data representation between systems:
– it is necessary to establish both syntactic (model-based) and semantic
interoperability to represent knowledge in a computable form
3
Cohort identification as part
Clinical trials life-cycle
Adapted from Source: Douglas Fridsma, MD, PhD
The University of Pittsburgh Cancer Institute
Centre for Pathology and Oncology Informatics
Clinical Trials Research Core
4
Query Formulation Workbench
•
Provides tools necessary to author, store and deploy queries of clinical data to
identify subjects for clinical studies:
– query authoring for the identification of research subjects based on existing
eHR data
– use of semantic mediator services
•
Semantically aware
•
Enables easy authoring of distributed searches to EHR and other clinical data
sources
•
Uses a controlled vocabulary service and appropriate standards-based
technological solutions
•
Automatically identifies ‘prevalent cases’ for research
– Count eligible subjects, flag the subjects for recruitment and consent by the local clinical
care team
– Full compliance with data protection legislation and best practice
5
Overall Design Approach
• The development of the system adopts a model-based approach, where
the TRANSFoRm Clinical Research Information Model (CRIM) provides a
computable information model for eligibility criteria.
• Criteria concepts, especially clinical concepts, can be browsed and
selected through the TRANSFoRm Integrated Vocabulary Service.
– The vocabulary service provides mappings from standard UMLS
concepts to standard EHR or clinical data sources’ coding schemes.
• The eligibility criteria are captured in a computable representation, based
on the Clinical Data Integration Model (CDIM) ontology
– CDIM captures an extensible common representation of clinical care
data
6
Conceptual Architecture
CDIM Ontology
l
)C
(1
ini
ca
o
lC
de
s
Vocabulary
Service
(2) Search Criteria
TRANSFoRm Query
Formulation Workbench
(5) Query Result
CDIM-DSM
mapping
Distributed Infrastructure
for Data Extraction and
Linkage
(3) Local Query
(4) Query Result
EHR
DS
Clinical Researcher
Provenance Service
7
Vocabulary Service: RCD v2/ICPC2
•
Read Codes (RCDv2) and International
Classification of Primary Care (ICPC2)
corpus of terms and their associated
mappings
– created to cater for the initial need of
the existence of specific primary care
oriented terminologies.
•
The UK NHS Connecting for Health
Terminology Centre - mappings from Read
Codes version 2 to SNOMED CT.
•
The Read Codes v2 database in
Transform VS is set up based on this
mapping so that Read Codes 2 concepts
can be linked to a UMLS search. Similar
approach for ICPC2.
•
ICPC2-ICD10 Thesaurus and mappings Transition Project @ University of
Amsterdam
•
The TRANSFoRm team is updating the
ICPC-ICD 10 mapping and Thesaurus
UMLS
Metathesaurus
SNOMED CT
Codes
Read Codes v2
Codes
UMLS
Metathesaurus
ICD-10
Thesaurus/Codes
ICPC2
Codes
8
Demo of the Vocabulary Service
9
Demo of Eligibility Criteria Creation
10
Submitting Queries
11
Identifying
prevalent
cases
through
eligible
counts
12
TRANSFoRm
Dr Theodoros N. Arvanitis
University of Birmingham
Birmingham Children’s Hospital NHS Trust
University of Birmingham Clinical Informatics Research Team:
Sarah Lim Choi Keung, James Rossiter, Lei Zhao