eCite Digital Repository

MLM-based automated query generation for CDSS evidence support


Afzal, M and Hussain, M and Ali, T and Khan, WA and Lee, S and Kang, BH, MLM-based automated query generation for CDSS evidence support, Lecture Notes in Computer Science 8867: Proceedings of the 8th International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI 2014), 2-5 December 2014, Belfast, UK, pp. 296-299. ISSN 0302-9743 (2014) [Refereed Conference Paper]

Copyright Statement

Copyright 2014 Springer International Publishing Switzerland

DOI: doi:10.1007/978-3-319-13102-3_49


Clinical decision support system (CDSS) is fast becoming a requirement in diverse medical domains to assist physicians in clinical decisions. Physicians look at the research evidences for satisfaction in CDSS assisted clinical decisions and also to keep their knowledge up-to-date. Research evidences are available in the form of studies, summaries, and other formats published in credible journals, books and reviews as online sources. The most important and critical part to get the evidences in a better way is the search query generation and its optimization. A query that is characterized by domain context and clinical workflow, and optimized for the target search engine in order to generate right and relevant results. In most cases, the search queries are generated manually, which require a lot of physicians’ time to get the right information. Other follow automated way of generating queries from electronic medical records, which make it difficult to associate evidences to the clinical decisions. The role of the source from where the queries are created is highly important. We are presenting the work of query generation from Medical Logic Modules (MLMs) as a main source of query contents. We create different query set from the concepts used in MLMs expended with domain ontology derived from SNOMED CT. The results are compiled with respect to coverage using classified training set of over 380 research articles. The proposed work is demonstrated to physicians and their feedback upon time saving as well as presentation of information in the context was highly positive.

Item Details

Item Type:Refereed Conference Paper
Keywords:MLM, CDSS evidence support
Research Division:Information and Computing Sciences
Research Group:Information systems
Research Field:Information systems not elsewhere classified
Objective Division:Information and Communication Services
Objective Group:Information services
Objective Field:Information services not elsewhere classified
UTAS Author:Kang, BH (Professor Byeong Kang)
ID Code:98421
Year Published:2014
Deposited By:Information and Communication Technology
Deposited On:2015-02-13
Last Modified:2018-03-28

Repository Staff Only: item control page