ThinkMiners: Disorder Recognition using Conditional Random Fields and Distributional Semantics

Proceedings of the 8th International Workshop on Semantic Evaluation (SemEval 2014), pages 652–656, Dublin, Ireland, August 23-24, 2014. ThinkMiners: Disorder Recognition using Conditional Random Fields and Distributional Semantics Ankur Parikh Avinesh PVS Joy Mustafi Lalit Agarwalla Ashish Mungi IBM India Pvt Ltd, IBM Software Group, Watson {anparikh,pavinesh,jmustafi,lalit.agarwalla,r1amungi}@in.ibm.com Abstract In 2014, SemEval organized multiple chal- lenges on natural language processing and information retrieval. One of the task was analysis of the clinical text. This challenge is further divided into two tasks. The task A of the challenge was to extract disor- der mention spans in the clinical text and the task B was to map each of the disor- der mentions to a unique Unified Medical Language System Concept Unique Iden- tifier. We participated in the task A and developed a clinical disorder recognition system. The proposed system consists of a Conditional Random Fields based ap- proach to recognize disorder entities. The SemEval challenge organizers manually annotated disorder entities in 298 clini- cal notes, of which 199 notes were used for training and 99 for development. On the test data, our system achieved the F- measure of 0.844 for entity recognition in relaxed and 0.689 in strict evaluation. Keywords: medical language processing, clinical concept extraction, conditional random fields. 1 Introduction Mining concepts from the electronic medical records such as clinical reports, discharge sum- maries as well as large number of doctor’s notes has become an utmost important task for auto- matic analysis in the medical domain. Identifica- tion and mapping of the concepts like symptoms, disorders, surgical procedures, body sites to a nor- malized standards are usually the first steps to- This work is licensed under a Creative Commons At- tribution 4.0 International Licence. Page numbers and pro- ceedings footer are added by the organisers. Licence details: http://creativecommons.org/licenses/by/4.0/ wards understanding natural language text in the medical records. In this paper, we describe a machine learning based disorder recognition system for the Task 7A of 2014 SemEval challenge. In Section 2 we give a background of the existing solutions to tackle the problem. Section 3 covers our approach in detail, followed by evaluation and conclusion in Section 4 and Section 5 respectively. 2 Background In recent times, many systems have been de- veloped to extract clinical concepts from vari- ous types of clinical notes. The earlier nat- ural language processing (NLP) systems were mainly built heavily using domain knowledge i.e. medical dictionaries. These systems in- clude MetaMap (Aronson and Lang, 2010), Hi- TEX (Zeng et al., 2006), KnowledgeMap (Denny et al., 2003), MedLEE (Friedman et al., 1994), SymText (Koehler, 1994) and Mplus (Christensen et al., 2002). In the past couple of years, re- searchers have been exploring the use of machine learning algorithms in the clinical concept detec- tion. To promote the research in this field many or- ganizations such as ShARe/CLEF, SemEval have organized a few clinical NLP challenges. In CLEF 2013 (Pradhan et al., 2013), the challenge was to recognize medication-related concepts. Both rule- based (Fan et al., 2013; Ramanan …

Content truncated to 500 words