IR in Biomedicine 2007

-- CALL FOR CHAPTERS -- Submission Deadline: November 1, 2007 Information Retrieval in Biomedicine: Natural Language Processing for Knowledge Integration A book edited by Pr. Violaine Prince and Dr Mathieu Roche, University of Montpellier and LIRMM-CNRS, France

Introduction

There is nowadays an intense interest for bio natural language processing. This field addresses the particular applications of natural language processing (NLP) to biological and medical areas. Naming such a set of applications denotes both the impact of NLP on the application domain. As a feedback, the peculiarities of the later seems to have made NLP evolve in a distinct and particular way.

Several articles and books chapters have been recently written on the subject (Ibekwe-Sanjuan 2007, Ananiadou and McNaught 2006, Scherf et al. 2005, Cohen and Hersh 2005 are among the latest...). The issue they tackle rises from the intensive research and publication activity in the medical area. A bibliographical database such as Medline contains several millions of articles and is thoroughly updated every day. Many medical researchers and practitioners need to read papers not only in their discipline but in other fields with which they have an interaction. For instance, cancer specialists need to browse papers in oncogenetics, radiology, chemistry, cellular biology, surgery and so forth.Every day new cross-studies are published, and the medical community cannot cope with such a high rate of information without being supported by automated or semi-automated tools in Information Retrieval and Knowledge Integration.

According to Swanson's pioneer work in the domain (Swanson 1986), the abundant medical data could be used as a hypothesis generator for orienting medical research. Since human operators cannot browse the huge amount of information, he suggested that hidden correlations could be automatically or semi-automatically found in this data, so as to suggest new tracks for investigation. Nowadays, the most recent works in text mining are able to suggest this type of scientific discovery: A recent work by Chun et al. (2006) shows that mining Medline abstracts brought up interesting topical relations between prostate cancer and genes. Beyond medicine, it is the whole field of the "living sciences", including all facets of biology, that might benefit from text mining methods and achievements (A recent paper by Ananiadou et al. (2006) describes perspective actions of text mining in systems biology).

The Overall Objective of the Book

In the fields of bio NLP there exists a need for an edited collection of articles in this area. Until now, the most intensively explored NLP areas in biomedicine are those related to lexical knowledge and terminology. Named entities recognition, abbreviations understanding and expansion, terminological knowledge management have been largely addressed, with more or less success. However, since NLP parsers are becoming more efficient, and word-based approaches have reached their limits, new trends, suggesting hybridation between linguistic knowledge and machine learning or statistics-based algorithms are being seriously investigated.

The book aims to provide relevant theoretical frameworks and latest empirical research findings in the area, according to a linguistic granularity. At the lexical and terminological levels, it aims at presenting original applications, going beyond the existing published work. At the sentence level, it should present the latest achievements, particularly by using NLP parsers. At the text/paragraph level, it is the relationship with topics and pragmatics that opens the road for a broader use of NLP in biomedicine. Moreover, two chapters will focus on aspects of NLP which are becoming crucial: Evaluation and Innovative Software.

The Target Audience: Professionals, PhD students and researchers working in the field of Text Mining, BioNLP, Medical Sciences, and Computer-Assisted Medical information systems. It is also relevant for computational linguists and linguists who want to solve particular problems brought out by the application domain. Moreover, the book will provide insights and support executives concerned with the management of expertise, knowledge, and information in health systems and biological textbases.

Recommended topics include the following:

Lexical-terminological level: Lexicology and terminology in BioNLP ; Using BIO ontologies within a language context ; Updating ontologies in biology or medicine with lexical knowledge

Sentence level: The question-answer approach in biomedicine; Operative knowledge derived from NLP parsing and/or semantic representation (application to biology and/or medicine); Approaches linking sentence level with either terminology or segment level

Segment level: A topical and topic change approach to BioNLP (for Information Retrieval or Knowledge Integration); Rhetorical structures, scripts, or other models of this granularity; Approaches involving language pragmatics in Biomedicine

Evaluation: Models or points of view in evaluating NLP approaches to biomedicine

Innovative Software in BioNLP (short papers)

SUBMISSION PROCEDURE

Researchers and practitioners are invited to submit on or before November 1, 2007, a 2-5 page manuscript proposal clearly explaining the mission and concerns of the proposed chapter. Authors of accepted proposals will be notified by December 1, 2007 about the status of their proposals and sent chapter organizational guidelines. Full chapters are expected to be submitted by March 15, 2008. All submitted chapters will be reviewed on a double-blind review basis. The book is scheduled to be published by IGI Global, www.igi-pub.com, publisher of the IGI Publishing (formerly Idea Group Publishing), Information Science Publishing, IRM Press, CyberTech Publishing and Information Science Reference (formerly Idea Group Reference) imprints.

Inquiries and submissions can be forwarded electronically (Word document) or by mail to: Pr Violaine Prince University of Montpellier 2 and LIRMM-CNRS 161 Ada Street F34392 Montpellier cedex 5 FRANCE Tel.: +334 67 41 86 74 Fax: +334 67 41 85 00 GSM: +336 07 34 01 00 E-mail: prince@lirmm.fr	 This CfP was obtained from WikiCFP