Usability-driven pruning of large ontologies: the case of SNOMED CT

López-García, Pablo; Boeker, Martin; Illarramendi, Arantza; Schulz, Stefan

doi:10.1136/amiajnl-2011-000503

Cited by 10 publications

(17 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In considering networks, the common nomenclature (UMLS) is a highly connected network ideal for finding relatedness between professions. A different network, Systematized Nomenclature of Medicine-Clinical Terms (SNOMED-CT) [51,52], has relationships designed as an a-cyclical tree, which would enable easy comparison between two terms but would increase the distance between any two terms. The full network of UMLS is more representative of the complex nature of biomedicine.…”

Section: Discussionmentioning

confidence: 99%

Physician nurse care: A new use of UMLS to measure professional contribution

Boyd

Lopez

Lugaresi

et al. 2018

International Journal of Medical Informatics

View full text Add to dashboard Cite

Background Physician and nurses have worked together for generations; however, their language and training are vastly different; comparing and contrasting their work and their joint impact on patient outcomes is difficult in light of this difference. At the same time, the EHR only includes the physician perspective via the physician-authored discharge summary, but not nurse documentation. Prior research in this area has focused on collaboration and the usage of similar terminology. Objective The objective of the study is to gain insight into interprofessional care by developing a computational metric to identify similarities, related concepts and differences in physician and nurse work. Methods 58 physician discharge summaries and the corresponding nurse plans of care were transformed into Unified Medical Language System (UMLS) Concept Unique Identifiers (CUIs). MedLEE, a Natural Language Processing (NLP) program, extracted “physician terms” from free-text physician summaries. The nursing plans of care were constructed using the HANDS© nursing documentation software. HANDS© utilizes structured terminologies: nursing diagnosis (NANDA-I), outcomes (NOC), and interventions (NIC) to create “nursing terms”. The physician’s and nurse’s terms were compared using the UMLS network for relatedness, overlaying the physician and nurse terms for comparison. Our overarching goal is to provide insight into the care, by innovatively applying graph algorithms to the UMLS network. We reveal the relationships between the care provided by each professional that is specific to the patient level. Results We found that only 26% of patients had synonyms (identical UMLS CUIs) between the two professions’ documentation. On average, physicians’ discharge summaries contain 27 terms and nurses’ documentation, 18. Traversing the UMLS network, we found an average of 4 terms related (distance less than 2) between the professions, leaving most concepts as unrelated between nurse and physician care. Conclusion Our hypothesis that physician’s and nurse’s practice domains are markedly different is supported by the preliminary, quantitative evidence we found. Leveraging the UMLS network and graph traversal algorithms, allows us to compare and contrast nursing and physician care on a single patient, enabling a more complete picture of patient care. We can differentiate professional contributions to patient outcomes and related and divergent concepts by each profession.

show abstract

Section: Discussionmentioning

confidence: 99%

Physician nurse care: A new use of UMLS to measure professional contribution

Boyd

Lopez

Lugaresi

et al. 2018

International Journal of Medical Informatics

View full text Add to dashboard Cite

show abstract

“…However, this may go against government or AMA policies. An alternative straightforward approach could be to conduct double coding (ICD-9-CM and ICD-10-CM) for the entangled ICD-9-CM codes and compare motifs in ICD-9-CM and ICD-10-CM in the final reports of the medical system or clinics, such as graph-pruning strategies to subsets offering reasonable coverage 24. However, dual coding is cost-prohibitive as coding to ICD-10-CM codes may require additional patient information that is available in patient charts but unobtainable from the historical ICD-9-CM claims.…”

Section: Discussionmentioning

confidence: 99%

The discriminatory cost of ICD-10-CM transition between clinical specialties: metrics, case study, and mitigating tools

Boyd

Burton

et al. 2013

J Am Med Inform Assoc

View full text Add to dashboard Cite

ObjectiveApplying the science of networks to quantify the discriminatory impact of the ICD-9-CM to ICD-10-CM transition between clinical specialties.Materials and MethodsDatasets were the Center for Medicaid and Medicare Services ICD-9-CM to ICD-10-CM mapping files, general equivalence mappings, and statewide Medicaid emergency department billing. Diagnoses were represented as nodes and their mappings as directional relationships. The complex network was synthesized as an aggregate of simpler motifs and tabulation per clinical specialty.ResultsWe identified five mapping motif categories: identity, class-to-subclass, subclass-to-class, convoluted, and no mapping. Convoluted mappings indicate that multiple ICD-9-CM and ICD-10-CM codes share complex, entangled, and non-reciprocal mappings. The proportions of convoluted diagnoses mappings (36% overall) range from 5% (hematology) to 60% (obstetrics and injuries). In a case study of 24 008 patient visits in 217 emergency departments, 27% of the costs are associated with convoluted diagnoses, with ‘abdominal pain’ and ‘gastroenteritis’ accounting for approximately 3.5%.DiscussionPrevious qualitative studies report that administrators and clinicians are likely to be challenged in understanding and managing their practice because of the ICD-10-CM transition. We substantiate the complexity of this transition with a thorough quantitative summary per clinical specialty, a case study, and the tools to apply this methodology easily to any clinical practice in the form of a web portal and analytic tables.ConclusionsPost-transition, successful management of frequent diseases with convoluted mapping network patterns is critical. The http://lussierlab.org/transition-to-ICD10CM web portal provides insight in linking onerous diseases to the ICD-10 transition.

show abstract

“…Ontology modularization techniques include graph-traversal [11–14] and logic-based techniques [15,16], whose extraction strategies depend on the ontology’s topology and its definitional axioms, respectively. The requirement of preserving the original ontology entailments, key to ontology modularization, however, adds a large number of terms to the module that are unlikely to be found in clinical documents, necessarily affecting precision and performance [17]. …”

Section: Introductionmentioning

confidence: 99%

“…One of the most relevant examples of such a subset is the CORE problem list subset of SNOMED CT (CORE) [20], which is only 1.50% the size of SNOMED CT but covers over 90% of the diagnoses and problem lists found in existing reference datasets. Public authoritative medical corpora are very scarce (with the notable exception of the Multiparameter Intelligent Monitoring in Intensive Care II clinical database [21]), and using a generalist corpus (e.g., MEDLINE) might not provide good enough results because of potential mismatch between content and vocabulary used in scientific abstracts and clinical jargon [17]. To extract the CORE subset, seven large-scale health care institutions collaborated to analyze their datasets.…”

Section: Introductionmentioning

confidence: 99%

Cross-domain targeted ontology subsets for annotation: The case of SNOMED CORE and RxNorm

López-García

LePendu

Musen

et al. 2014

Journal of Biomedical Informatics

Self Cite

View full text Add to dashboard Cite

The benefits of using ontology subsets versus full ontologies are well-documented for many applications. In this study, we propose an efficient subset extraction approach for a domain using a biomedical ontology repository with mappings, a cross-ontology, and a source subset from a related domain. As a case study, we extracted a subset of drugs from RxNorm using the UMLS Metathesaurus, the NDF-RT cross-ontology, and the CORE problem list subset of SNOMED CT. The extracted subset, which we termed RxNorm/CORE, was 4% the size of the full RxNorm (0.4% when considering ingredients only). For evaluation, we used CORE and RxNorm/CORE as thesauri for the annotation of clinical documents and compared their performance to that of their respective full ontologies (i.e., SNOMED CT and RxNorm). The wide range in recall of both CORE (29–69%) and RxNorm/CORE (21–35%) suggests that more quantitative research is needed to assess the benefits of using ontology subsets as thesauri in annotation applications. Our approach to subset extraction, however, opens a door to help create other types of clinically useful domain specific subsets and acts as an alternative in scenarios where well-established subset extraction techniques might suffer from difficulties or cannot be applied.

show abstract

Usability-driven pruning of large ontologies: the case of SNOMED CT

Abstract: Graph-traversal strategies and frequency data from an authoritative source can prune large biomedical ontologies and produce useful subsets that still exhibit acceptable coverage. However, a clinical corpus closer to the specific use case is preferred when available.

Cited by 10 publications

References 11 publications

Physician nurse care: A new use of UMLS to measure professional contribution

Physician nurse care: A new use of UMLS to measure professional contribution

The discriminatory cost of ICD-10-CM transition between clinical specialties: metrics, case study, and mitigating tools

Cross-domain targeted ontology subsets for annotation: The case of SNOMED CORE and RxNorm

Contact Info

Product

Resources

About