Learning objectives
The goal of this project was to calculate medical outcomes from disparate sources of information (Fig. 1), but to achieve this aim an ontological structure had to be created to efficiently relate different types of data (Fig. 2). The challenge is that information in an electronic medical record (EMR) is often coded with multiple ontologies (controlled vocabularies with defined relationships between terms) depending on the medical discipline and purpose. The lack of connection and interoperability between ontologies makes it difficult to correlate information in order...
Background
Our goal was to calculate medical outcomes using disparate sources of information, butit was necessary to create an ontological structure to relate different sources of information coded with various ontologies.
An ontology is a controlled vocabulary with defined relationships between terms. Hundreds of biomedical ontologies exist that are designed for different purposes and audiences. The National Center for Biomedical Ontology currently maintains a repository of 828 ontologies.2
Common ontologies used in clinical medicine and their purposes include:
SNOMED-CT - medical diagnoses and findings3
CPT -...
Findings and procedure details
Several innovations were employed to achieve the goal of calculating medical outcomes:
1. Structured reporting solution that integrates related information in timelines
The newontology development began during the creation of a multimedia structured reporting application that required medical terminology to tag image findings. The reporting solution works by (1) recording key images, (2) transcribing a radiologist’s verbal description to text, (3) tagging the information with metadata referenced to the ontology using natural language processing (NLP), and (4) assembling a multimedia report with related radiological findings...
Conclusion
Relating radiological findings to other medical events is essential for the calculation of medical outcomes, but this task requires the use of a multidisciplinary ontology to relatedifferent sources of information.
Personal information and conflict of interest
D. J. Vining; Houston, TX/US - Founder at VisionSR A. Prisacariu; Houston/US - Owner at Patrisoft Outsourcing M. Kontak; Houston/US - Consultant at VisionSR A. Pitici; Houston/US - Employee at Patrisoft Outsourcing C. Popovici; Houston/US - Employee at Patrisoft Outsourcing
References
Amith M, He Z, Bian J, Lossio-Ventura JA, Tao C. Assessing the practice of biomedical ontology evaluation: Gaps and opportunities. J Biomed Inform 2018;80:1-13
Welcome to BioPortal, the world's most comprehensive repository of biomedical ontologies. BioPortal, The Board of Trustees of Leland Stanford Junior University, 2019, http://bioportal.bioontology.org/.
SNOMED International determines global standards for health terms, an essential part of improving the health of humankind. SNOMED International, SNOMED International, 2020, http://www.snomed.org/.
CPT. AMA, American Medical Association, 2020, https://www.ama-assn.org/practice-management/cpt.
RadLex radiology lexicon. RSNA, Radiological Society of North...