A study of structured clinical abstracts and the semantic classification of sentences

Grace Y. Chung, Enrico Coiera

Research output: Chapter in Book/Report/Conference proceedingConference proceeding contribution

15 Citations (Scopus)

Abstract

This paper describes experiments in classifying sentences of medical abstracts into a number of semantic classes given by section headings in structured abstracts. Using conditional random fields, we obtain -scores ranging from 0.72 to 0.97. By using a small set of sentences that appear under the PAR-TICPANTS heading, we demonstrate that it is possible to recognize sentences that describe population characteristics of a study. We present a detailed study of the structure of abstracts of randomized clinical trials, and examine how sentences labeled under PARTICIPANTS could be used to summarize the population group.

Original languageEnglish
Title of host publicationBioNLP '07 Proceedings of the Workshop on BioNLP 2007
Subtitle of host publicationBiological, Translational, and Clinical Language Processing
Place of PublicationStroudsburg, PA, USA
PublisherAssociation for Computational Linguistics (ACL)
Pages121-128
Number of pages8
Publication statusPublished - 2007
Externally publishedYes
EventACL 2007 Workshop on Biological, Translational, and Clinical Language Processing, BioNLP 2007 - Prague, Czech Republic
Duration: 29 Jun 2007 → …

Conference

ConferenceACL 2007 Workshop on Biological, Translational, and Clinical Language Processing, BioNLP 2007
CountryCzech Republic
CityPrague
Period29/06/07 → …

Fingerprint Dive into the research topics of 'A study of structured clinical abstracts and the semantic classification of sentences'. Together they form a unique fingerprint.

  • Cite this

    Chung, G. Y., & Coiera, E. (2007). A study of structured clinical abstracts and the semantic classification of sentences. In BioNLP '07 Proceedings of the Workshop on BioNLP 2007: Biological, Translational, and Clinical Language Processing (pp. 121-128). Stroudsburg, PA, USA: Association for Computational Linguistics (ACL).