reusing electronic dental record (edr) data through an ontology
DESCRIPTION
Objective: A key question for healthcare is how to operationalize the vision of the Learning Healthcare System, in which electronic health record data become a continuous information source for quality assurance and research. The development of evidence, comparative effectiveness research, and clinical studies require analysis of data generated during clinical practice. These data are increasingly stored in electronic dental records (EDR). The purpose of this project is to develop a generalizable method for extracting and analyzing research data from EDRs. Method: In a highly iterative and collaborative process, our team of dentists, informaticians, ontologists and clinical dental researchers developed a pilot process for extracting clinical data from Eaglesoft, a commercial EDR. We defined a set of research questions focused on restorative dentistry; constructed the Oral Health and Disease Ontology (OHD) to represent the study variables; developed mappings for the data; and created a knowledge base that used the OHD to represent clinical data. Result: We represented data from about 4,500 patients from a single dental practice in our knowledge base. We extracted 232,270 clinical records, of which about 54,000 documented restorative, endodontic and surgical procedures. Currently, the OHD includes 213 classes and reuses 1,658 classes from other ontologies. Most of the classes created de novo were related to procedures, visits, exams and CDT codes that were not available in existing biomedical ontologies. We have developed an initial set of queries to extract data about patients, teeth, surfaces, restorations, and findings, and are currently in the process of conducting initial statistical analyses of the data. Conclusion: Our study is a novel application of an ontology-based approach to develop a generalizable method for extracting and reusing data from EDRs. Further work will establish a complete, open and reproducible workflow for reusing data from a variety of EDRs for research and quality assurance. This abstract is based on research that was funded entirely or partially by an outside source: National Institute of Dental Craniofacial Research (NIDCR) 5R21DE019683-02 and National Institute of Dental Craniofacial Research (NIDCR) 1R21DE021178-01A1.TRANSCRIPT
Reusing Electronic Dental Record Data Through an Ontology Titus Schleyer1, DMD, PhD; Alan Ruttenberg2, MS; William Duncan2, MS; Melissa Haendel3, PhD; Carlo Torniai3, PhD; Amit Acharya4, PhD; Mei Song1, PhD; Thankam Thyvalikakath1, DMD, PhD;
Kaihong Liu1, PhD; Pedro Hernandez5, DMD, MS 1University of Pittsburgh, PA; 2University of Buffalo, NY; 3Oregon Health & Science University, OR; 4Marshfield Clinical Research Foundation, WI; 5Reparto Universitario, PR
University of Pittsburgh Department of Biomedical Informatics
Background
• key question in healthcare: how can we operationalize the vision of the Learning Healthcare System?
• Electronic health record data could be a continuous information source for quality assurance and research.
• Clinical data need to be analyzed to develop evidence and conduct comparative effectiveness research.
• Clinical data are increasingly stored in electronic dental records (EDR)1.
Objective
develop a generalizable method for extracting and analyzing research data from EDRs.
Methods
• develop a pilot process for data extraction from Eaglesoft
• define a set of research questions focused on restorative dentistry
• construct the Oral Health and Disease Ontology (OHD) to represent study variables
• map data from a Eaglesoft patient database to the OHD
• create a knowledge base to represent clinical data using the OHD
Results
• ˜4,500 patients data represented• 232,270 clinical records extracted • 213 new classes created in OHD• 1,658 classes reused from other ontologies• a set of queries to extract data developed
Conclusion
• novel application of an ontology-based approach to extract data from EDRs
• future research to establish a complete, open and reproducible workflow for data extraction and analysis across EDRs
Contact information
http://di.dental.pitt.edu
Email: [email protected]
Phone: 412-648-8886
This research is supported by the NIH grants R21-DE-19683 and R21-DE-21178
Perform Statistical Analyses
Import data
Query aggregated and classified data
Classify data
TriplestoreRepositoryTriplestoreRepository
Electronic Dental Record SystemsElectronic Dental Record Systems
Oral Health and Disease OntologyOral Health and Disease Ontology
1.Schleyer et al. Electronic dental record use and clinical information management patterns among practitioner-investigators in The Dental Practice-Based Research Network. JADA 2013.
2.Schleyer et al. An ontology-based method for secondary use of electronic dental record data. AMIA Summit on Clinical Research Informatics (CRI) 2013.