Summarization from medical documents: a survey

Abstract

Objective: The aim of this paper is to survey the recent work in medical documents summarization. Background: During the last decade, documents summarization got increasing attention by the AI research community. More recently it also attracted the interest of the medical research community as well, due to the enormous growth of information that is available to the physicians and researchers in medicine, through the large and growing number of published journals, conference proceedings, medical sites and portals on the World Wide Web, electronic medical records, etc. Methodology: This survey gives first a general background on documents summarization, presenting the factors that summarization depends upon, discussing evaluation issues and describing briefly the various types of summarization techniques. It then examines the characteristics of the medical domain through the different types of medical documents. Finally, it presents and discusses the summarization techniques used so far in the medical domain, referring to the corresponding systems and their characteristics. Discussion and Conclusions: The paper discusses thoroughly the promising paths for future research in medical documents summarization. It mainly focuses on the issue of scaling to large collections of documents in various languages and from different media, on personalization issues, on portability to new sub-domains, and on the integration of summarization technology in practical applications.

An Investigation Concerning the Generation of Text Summarisation Classifiers Using Secondary Data

“…Jones et al [14] proposed that the key factors involved in the summarisation of text be categorised according to: (i) the input, (ii) the purpose of the summarisation and (iii) the desired output (this last category is of course related to the second). Afantenos et al [1] provided a more "fine grained" categorisation founded on that of Jones et al The input factors considered were: (i) the number of documents used (singledocument or multi-document), (ii) the data format in which the documents are presented (text or multimedia) and (iii) the language or languages in which the text was written (monolingual, multilingual and cross-lingual). The purpose factors were sub-categorised in terms of: (i) the nature of the text (indicative or informative), (ii) how specific the summary must be for the intended audience (generic or user oriented) and (iii) how specific the summary must be in terms of the domain or field of study (general purpose or domain specific).…”

Syntactic sentence compression in the biomedical domain: facilitating access to related articles

“…This work explores a more sophisticated solution based on linguistic analysis-we hypothesize that sentence compression techniques can potentially deliver the best of both worlds: by shortening article titles in a linguistically meaningful way, the interface could deliver much of the same substance in a smaller amount of space. Ideally, the output of a compression algorithm should be both indicative and informativeproperties often discussed in the context of document summarization (Afantenos et al 2005). Indicative summaries suggest the content of an underlying information object without necessarily giving away details-often the aim is to entice users, or at least alert them to the presence of a particular information object.…”

“…Although document summarization techniques have principally been applied to newswire text, there is a body of research that deals specifically with the summarization of medical documents-see (Afantenos et al 2005) for a survey. A noteworthy example is PERSIVAL (McKeown et al 2003;Elhadad et al 2005), which leverages patient records to generate personalized summaries.…”

Evaluating Web Search Result Summaries

“…Methods of evaluating automatic summarisation systems can be broadly classified into two types: intrinsic and extrinsic [1] [10]. Intrinsic evaluation assesses the quality of a summary per se, examining aspects such as coherence, readability, grammaticality, and fidelity.…”