Welcome to the InfoSci Platform
Reference Hub7
Medical Case Based Reasoning Frameworks: Current Developments and Future Directions

Medical Case Based Reasoning Frameworks: Current Developments and Future Directions

Shaker El-Sappagh, Mohammed Mahfouz Elmogy
Copyright: © 2016 |Volume: 8 |Issue: 3 |Pages: 32
ISSN: 1941-6296|EISSN: 1941-630X|EISBN13: 9781466690547|DOI: 10.4018/IJDSST.2016070103
Cite Article Cite Article

MLA

El-Sappagh, Shaker, and Mohammed Mahfouz Elmogy. "Medical Case Based Reasoning Frameworks: Current Developments and Future Directions." IJDSST vol.8, no.3 2016: pp.31-62. http://doi.org/10.4018/IJDSST.2016070103

APA

El-Sappagh, S. & Elmogy, M. M. (2016). Medical Case Based Reasoning Frameworks: Current Developments and Future Directions. International Journal of Decision Support System Technology (IJDSST), 8(3), 31-62. http://doi.org/10.4018/IJDSST.2016070103

Chicago

El-Sappagh, Shaker, and Mohammed Mahfouz Elmogy. "Medical Case Based Reasoning Frameworks: Current Developments and Future Directions," International Journal of Decision Support System Technology (IJDSST) 8, no.3: 31-62. http://doi.org/10.4018/IJDSST.2016070103

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

Case-Based Reasoning (CBR) is one of the most suitable AI techniques for building clinical decision support systems. Medical domain complexity introduces many challenges for building these systems. Building the systems' knowledge base from the Electronic Health Record (EHR), the encoding of case-base knowledge with standard medical ontology, and the handling of vague data are examples of these challenges. Although several advantages of using CBR in medicine have been identified, there are no real systems acceptable to physicians. This systematic review examines the current state of CBR and its limitations in the medical domain, especially for diabetes mellitus. The critical evaluation of the status of diabetes CBR systems presents unique opportunities for improving these systems. The literature review covers most of the English language studies extracted from relevant databases by using search terms relating CBR, ontology, Fuzzy, and standard terminology concepts. The authors identify 38 articles published between 1999 and 15 January 2015, which represent original researches in CBR domain. The study includes 15 (39.5%) non-medical studies and 23 (60.5%) medical studies with ~22% for diabetes CBR. A list of 18 integrated evaluation metrics has been proposed and used in the analysis. The results show that the non-medical CBR systems achieved higher advances (50%) than medical systems (42.9%). In addition, the diabetes management CBR systems achieve the lowest advances (21.4%) compared to other systems. These shortages explain the question “why CBR paradigm are not fully utilized in the commercial medical systems?” As a result, there is a distinct need for more comprehensive enhancements in clinical CBR especially diabetes systems.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.