=Paper=
{{Paper
|id=Vol-1327/30
|storemode=property
|title=An Ontologic Approach to Leverage Surgical Training Data Development and Application of a CranioMaxilloFacial Ontology
|pdfUrl=https://ceur-ws.org/Vol-1327/icbo2014_paper_37.pdf
|volume=Vol-1327
|dblpUrl=https://dblp.org/rec/conf/icbo/GroveE14
}}
==An Ontologic Approach to Leverage Surgical Training Data Development and Application of a CranioMaxilloFacial Ontology==
https://ceur-ws.org/Vol-1327/icbo2014_paper_37.pdf
ICBO 2014 Proceedings
An Ontologic Approach to Leverage Surgical
Training Data
Development and application of a CranioMaxilloFacial ontology
Michael Grove
Jeff Emch
Oregon Health & Science University
Department of Medical Informatics & Clinical Epidemiology
Portland, Oregon, United States
grovmi@ohsu.edu
emch@ohsu.edu
Oral and maxillofacial surgery (OMS) is a surgical specialty
involving procedures on the neck and head. OMS training II. CRANIOMAXILLOFACIAL ONTOLOGY
program accreditation and evaluation requires reporting the
surgical experiences of trainees. Current systems for tracking A. Ontology Development Model
these experiences are based on coarse payment coding systems. To develop the CMF ontology, we leveraged an existing
To provide more granular training data, we are developing an clinical ontology (SNOMED-CT) and built in enhancements to
ontology-driven surgical resident training log (OMSLog). We use improve domain knowledge representation. This was
a blended architecture consisting of an interface for trainees,
performed by a clinical domain expert who identified relevant
faculty, and/or administrators to record surgical experiences, a
clinical finding and procedure concepts for leverage in
traditional relational database back-end for data storage, with
both supported by a new domain ontology. The
SNOMED and extended these into more granular domain
CranioMaxilloFacial (CMF) ontology is built on a SNOMED CT concepts. The logic model of the CMF ontology is consistent
foundation and is extended to include granular domain concepts with SNOMED as it arranges concept terminology into the
and educational experiences. Current results are a pilot existing SNOMED hierarchy and leverages SNOMED’s
graphical user interface (GUI) driven by a >7,000 concept property types and anatomical sites. Additionally, a custom
domain ontology. Future steps include pilot testing in the class hierarchy of OMS educational concepts was created to
residency program and ontological alignment with the Human characterize the educational experience of each respective
Phenotype Ontology (HPO). clinical finding and procedure concept. Concept definitions,
synonyms, and mappings to CPT, ICD-9, and ICD-10 will be
Keywords—Oral and maxillofacial surgery; craniofacial; included as annotations.
ontology; training; human phenotype ontology
B. System Architecture
I. INTRODUCTION The application interface (Figures 1 and 2) is built on an
Oral and Maxillofacial (OMS) residents are required to open-source Java web application stack utilizing Linux, the
track their surgical experiences in non-standardized program Apache webserver, with Java server pages (JSP) being hosted
training logs. Each year, OMS training programs expend by a by Tomcat server. All system code is sub-versioned using
resources to gather disparate data to meet reporting GitHub. Static data and reporting is supported by a MySQL
requirements for the Commission on Dental Accreditation relational database (RDB). The resource description framework
(CODA) Annual Survey. In both cases, localized, ad hoc tools (RDF) triple-store is indexed via Solr/Lucene to support rapid
populated with data based on clinical billing terminologies querying and traversal of the procedure and diagnosis trees for
such as the International Classification of Disease (ICD-9) and browsing and selection. Both the RDB and RDF are driven by
Current Procedure Terminology (CPT) are used. the CMF ontology. The ontology will be rendered and
versioned as a set of OWL files that merge to drive the
Reliance on reimbursement codes for clinical data yields SOLR/Lucene functionality. Next steps include mapping
coarse reporting that does not correlate educational experience historical log data to the existing RDB, and performing pilot
with surgical competency. Given the impediments of current testing and evaluation of the system. We are also developing a
OMS data format and flow, opportunities for transparent, real- plan to align the CMF ontology with the Human Phenotype
time individual and program-level quality improvement Ontology (HPO) in order to contribute new craniofacial
activities are being missed. To address this problem, we malformation classes to the existing HPO class hierarchy.
developed the “OMSLog,” a resident log system driven by a
new CranioMaxilloFacial (CMF) domain ontology.
Funding provided by NIH Grant number 2T15LM007088-21 REVISED
104
ICBO 2014 Proceedings
III. SCREENSHOTS
Figure 1 OMSLog Term Entry & Association
Figure 2 OMSLog Term Search Builder
105