Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2006:2006:56-60.

Generating executable knowledge for evidence-based medicine using natural language and semantic processing

Tara Borlawsky  1 , Carol Friedman, Yves A Lussier
Affiliations

Generating executable knowledge for evidence-based medicine using natural language and semantic processing

Tara Borlawsky et al. AMIA Annu Symp Proc. 2006.

Abstract

With an increase in the prevalence of patients having multiple medical conditions, along with the increasing number of medical information sources, an intelligent approach is required to integrate the answers to physicians' patient-related questions into clinical practice in the shortest, most specific way possible. Cochrane Scientific Reviews are currently considered to be the "gold standard" for evidence-based medicine (EBM), because of their well-defined systematic approach to assessing the available medical information. In order to develop semantic approaches for enabling the reuse of these Reviews, a system for producing executable knowledge was designed using a natural language processing (NLP) system we developed (BioMedLEE), and semantic processing techniques. Though BioMedLEE was not designed for or trained over the Cochrane Reviews, this study shows that disease, therapy and drug concepts can be extracted and correlated with an overall recall of 80.3%, coding precision of 94.1%, and concept-concept relationship precision of 87.3%.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Cochrane Review excerpt preparation and knowledge engineering methods.
Figure 2
Figure 2
Overview of EBM-TREE algorithm. AKE = Automated Knowledge Extraction; RP = Relation Processing; KO = Knowledge Object
Figure 3
Figure 3
Example of BioMedLEE output.
Figure 4
Figure 4
Recall of NLP and coding.
Figure 5
Figure 5
Precision of knowledge extraction.
See this image and copyright information in PMC

References

    1. Ebell MH, Barry HC. InfoRetriever: rapid access to evidence-based information on a hand-held computer. MD Comput. 1998;15(5):289, 292–7. - PubMed
    1. Delvenne C, Pasleau F. Organizing access to Evidence-Based Medicine resources on the Web. Comput Methods Programs Biomed. 2003;71(1):1–10. - PubMed
    1. Dong P, Mondry A. Enhanced quality and quantity of retrieval of Critically Appraised Topics using the CAT Crawler. Med Inform Internet Med. 2004;29(1):43–55. - PubMed
    1. Rindflesch TC, et al. Semantic relations asserting the etiology of genetic diseases. AMIA Annu Symp Proc. 2003:554–8. - PMC - PubMed
    1. Aronson AR. Effective Mapping of Biomedical Text to the UMLS Metathesaurus: The MetaMap Program. Proc AMIA Symp. 2001 - PMC - PubMed

Publication types

LinkOut - more resources