Building Large Collections of Chinese and English Medical Terms from Semi-Structured and Encyclopedia Websites

  • Yan Xu ,
  • Yining Wang ,
  • Jian-Tao Sun ,
  • ,
  • Junichi Tsujii ,
  • Eric Chang

PLOS ONE |

Funding: This work was supported by Microsoft Research Asia (MSR Asia). The work was also supported by MSRA eHealth grant, grant 61073077 from National Science Foundation of China and grant SKLSDE-2011ZX-13 from State Key Laboratory of Software Development Environment in Beihang University in China. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

To build large collections of medical terms from semi-structured information sources (e.g. tables, lists, etc.) and encyclopedia sites on the web. The terms are classified into the three semantic categories, Medical Problems, Medications, and Medical Tests, which were used in i2b2 challenge tasks. We developed two systems, one for Chinese and another for English terms. The two systems share the same methodology and use the same software with minimum language dependent parts. We produced large collections of terms by exploiting billions of semi-structured information sources and encyclopedia sites on the Web. The standard performance metric of recall (R) is extended to three different types of Recall to take the surface variability of terms into consideration. They are Surface Recall (), Object Recall (), and Surface Head recall (). We use two test sets for Chinese. For English, we use a collection of terms in the 2010 i2b2 text. Two collections of terms, one for English and the other for Chinese, have been created. The terms in these collections are classified as either of Medical Problems, Medications, or Medical Tests in the i2b2 challenge tasks. The English collection contains 49,249 (Problems), 89,591 (Medications) and 25,107 (Tests) terms, while the Chinese one contains 66,780 (Problems), 101,025 (Medications), and 15,032 (Tests) terms. The proposed method of constructing a large collection of medical terms is both efficient and effective, and, most of all, independent of language. The collections will be made publicly available.