Decision Support, Knowledge Representation and Management: Structuring Knowledge for Better Access

Self Cite

SummaryTo summarize excellent current research in the field of knowledge representation and management (KRM).A synopsis of the articles selected for the IMIA Yearbook 2011 is provided and an attempt to highlight the current trends in the field is sketched.This last decade, with the extension of the text-based web towards a semantic-structured web, NLP techniques have experienced a renewed interest in knowledge extraction. This trend is corroborated through the five papers selected for the KRM section of the Yearbook 2011. They all depict outstanding studies that exploit NLP technologies whenever possible in order to accurately extract meaningful information from various biomedical textual sources.Bringing semantic structure to the meaningful content of textual web pages affords the user with cooperative sharing and intelligent finding of electronic data. As exemplified by the best paper selection, more and more advanced biomedical applications aim at exploiting the meaningful richness of free-text documents in order to generate semantic metadata and recently to learn and populate domain ontologies. These later are becoming a key piece as they allow portraying the semantics of the Semantic Web content. Maintaining their consistency with documents and semantic annotations that refer to them is a crucial challenge of the Semantic Web for the coming years.

Section: Best Paper Selectionmentioning

confidence: 98%

Knowledge Representation and Management: Benefits and Challenges of the Semantic Web for the Fields of KRM and NLP

Rassinoux

2011

Self Cite

“…The paper of Johnson et al [12], already quoted in the IMIA Yearbook 2008 [13], proposes a design in which unstructured text and coded data are fused into a single model called structured narrative. A real time natural language processing module, coupled with an ontology of document types and sections, are used to analyze clinical documents and identify relevant concepts and their modifiers.…”

Section: Best Paper Selectionmentioning

confidence: 99%

Knowledge Representation and Management: Towards Interoperable Medical Terminologies

Rassinoux¹

2009

Self Cite

Summary Objective To summarize current outstanding research in the field of knowledge representation and management. Methods Synopsis of the articles selected for the IMIA Yearbook 2009. Results Four excellent papers have been selected for the section knowledge representation and management. All these papers are concerned with terminological systems whether it is to detect equivalent concept definitions, to structure narratives, to compare existing lay and professional terms or to reformulate the SNOMED’s logical formalism towards a more expressive language. Conclusion The ability of one computer system to access and use the resources of another system becomes crucial in a world where the amount of electronically stored data increases continuously. The selected papers for the section knowledge representation and management corroborate that interoperable medical terminologies play a growing and strategic role in Electronic Health Record (EHR) systems. Indeed, they allow data in clinical systems to be defined in a more uniform and granular manner, thus leading to flexible, semantically interoperable and trust-worthy health information systems.

“…"Building Capacity Worldwide" is the topic of the IMIA Yearbook 2010 and one way of contributing to this issue, in the field of knowledge representation and management, is to extend mechanisms to extract, collect, categorize and disseminate relevant information around the world. In order to reach this goal, it is necessary to add some sort of structured form to the significant content of the continually growing body of clinical and biomedical documents as well as web pages, as already highlighted in the IMIA Yearbook 2008 [7]. Two major international initiatives, showing valuable insights and guidance for structuring textual information, are worth noting.…”

Section: Introductionmentioning

confidence: 99%

Knowledge Representation and Management

2010

Self Cite

SummaryObjectives: To summarize current outstanding research in the field of knowledge representation and management. Method: Synopsis of the articles selected for the IMIA Yearbook 2010. Results: Four interesting papers, dealing with structured knowledge, have been selected for the section knowledge representation and management. Combining the newest techniques in computational linguistics and natural language processing with the latest methods in statistical data analysis, machine learning and text mining has proved to be efficient for turning unstructured textual information into meaningful knowledge. Three of the four selected papers for the section knowledge representation and management corroborate this approach and depict various experiments conducted to .extract meaningful knowledge from unstructured free texts such as extracting cancer disease characteristics from pathology reports, or extracting protein-protein interactions from biomedical papers, as well as extracting knowledge for the support of hypothesis generation in molecular biology from the Medline literature. Finally, the last paper addresses the level of formally representing and structuring information within clinical terminologies in order to render such information easily available and shareable among the health informatics community. Conclusions: Delivering common powerful tools able to automatically extract meaningful information from the huge amount of electronically unstructured free texts is an essential step towards promoting sharing and reusability across applications, domains, and institutions thus contributing to building capacities worldwide.