The PsyMine Corpus - A Corpus annotated with Psychiatric Disorders and their Etiological Factors

The PsyMine Corpus - A Corpus annotated with Psychiatric Disorders and their Etiological Factors Tilia Renate Ellendorff, Simon Foster, Fabio Rinaldi University of Zurich Institute of Computational Linguistics, Binzmühlestrasse 14, 8050 Zürich, Switzerland Epidemiology, Biostatistics and Prevention Institute (EBPI), Hirschengraben 84, 8001 Zürich, Switzerland tilia.ellendorff@uzh.ch, fabio.rinaldi@uzh.ch, simon.foster@uzh.ch Abstract We present the first version of a corpus annotated for psychiatric disorders and their etiological factors. The paper describes the choice of text, annotated entities and events/relations as well as the annotation scheme and procedure applied. The corpus is featuring a selection of focus psychiatric disorders including depressive disorder, anxiety disorder, obsessive-compulsive disorder, phobic disorders and panic disorder. Etiological factors for these focus disorders are widespread and include genetic, physiological, sociological and environmental factors among others. Etiological events, including annotated evidence text, represent the interactions between their focus disorders and their etiological factors. Additionally to these core events, symptomatic and treatment events have been annotated. The current version of the corpus includes 175 scientific abstracts. All entities and events/relations have been manually annotated by domain experts and scores of inter-annotator agreement are presented. The aim of the corpus is to provide a first gold standard to support the development of biomedical text mining applications for the specific area of mental disorders which belong to the main contributors to the contemporary burden of disease. Keywords: Information Extraction, Biomedical Text Mining, Corpus Annotation 1. Introduction Mental disorders are one of the main contributors to the contemporary burden of disease and one of the ma- jor groups of disorders that cause disability (Murray and Lopez, 2013). Therefore, it is important for research in the medical domain to advance in this field. One priority is to discover etiological factors (potential causes) of mental disorders. However, ongoing research faces the problem of having important information scattered across numerous resources, which are mostly textual. Furthermore, men- tal disorders can have a wide range of underlying factors, spanning over different areas, such as genetic factors, soci- ological factors, chemical factors and many more. Most of this information is encoded in the written text of research articles of the domain. This makes it hard for researchers to gain an overview of previously discovered causes and therefore obtain a complete picture of a mental disorder and its etiological factors. Biomedical text mining can help by automatically extracting relevant information from written text and presenting it in a structured, more easily acces- sible format. For building a reliable text mining system, annotated corpora are indispensable. Additionally, systems need to be trained on corpora of the same domain as the target domain in order to show a good performance. With the PsyMine corpus we present a corpus that is pri- marily meant to be used for evaluating but could also pos- sibly be used for building a biomedical text mining system for extracting mental disorders together with possible eti- ological factors. The annotation of the PsyMine is part of the PsyMine project1. In the following sections we will de- scribe the corpus, the …

Content truncated to 500 words