Format

Send to

Choose Destination
See comment in PubMed Commons below
Stud Health Technol Inform. 2017;235:271-275.

Personalized Guideline-Based Treatment Recommendations Using Natural Language Processing Techniques.

Author information

1
Department of Medical Informatics, University of Applied Sciences and Arts, Dortmund, Germany.

Abstract

Clinical guidelines and clinical pathways are accepted and proven instruments for quality assurance and process optimization. Today, electronic representation of clinical guidelines exists as unstructured text, but is not well-integrated with patient-specific information from electronic health records. Consequently, generic content of the clinical guidelines is accessible, but it is not possible to visualize the position of the patient on the clinical pathway, decision support cannot be provided by personalized guidelines for the next treatment step. The Systematized Nomenclature of Medicine - Clinical Terms (SNOMED CT) provides common reference terminology as well as the semantic link for combining the pathways and the patient-specific information. This paper proposes a model-based approach to support the development of guideline-compliant pathways combined with patient-specific structured and unstructured information using SNOMED CT. To identify SNOMED CT concepts, a software was developed to extract SNOMED CT codes out of structured and unstructured German data to map these with clinical pathways annotated in accordance with the systematized nomenclature.

KEYWORDS:

clinical guidelines; clinical pathways; information extraction; knowledge management; natural language processing

PMID:
28423796
PubMed Commons home

PubMed Commons

0 comments
How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for IOS Press
    Loading ...
    Support Center