Addis Ababa University Libraries Electronic Thesis and Dissertations: AAU-ETD! >
School of Information Science and Computer Science >
Thesis - Computer Science >

Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/2588

Authors: Elias, Muluneh
Advisors: Fitsum Admasu (PhD
Copyright: Oct-2009
Date Added: 4-May-2012
Publisher: AAU
Abstract: Due to advances in word processing technologies, spell checking and grammar checks have become ubiquitous. However, checking conceptual errors from any given text has received little attention from similar researches. An example for conceptual error is contradiction. The detection of conceptual errors on any plain text is a challenging task for computers. Two of the major challenges are the unstructured nature of text documents and the lack of representation of common sense and domain specific knowledge in machine understandable format. Researches show that thousands of people are killed per annum due to medication errors. Medication errors are mainly caused by the omission of facts about the patient‟s conditions or current drug intakes at the point of prescription. To tackle this problem, many hospitals have implemented computer physician order entry (CPOE) systems. CPOE systems are capable of detecting drug – drug interactions at the time of prescription. However, most information about the patient is stored in the form of textual narration in the medical records. Information contained in the narrated record about the patient is therefore not used by CPOE systems to validate the subjected medications. This thesis analyses sample medical notes to find patterns that could be explored towards using information encoded in plain medical notes for the detection of conceptual errors. The thesis proposes an ontology based architecture for a conceptual error detection system on medical notes. It also demonstrates results from the application of the prototype.
URI: http://hdl.handle.net/123456789/2588
Appears in:Thesis - Computer Science

Files in This Item:

File Description SizeFormat
1391483241784185837905578954351938600431.06 MBAdobe PDFView/Open

Items in the AAUL Digital Library are protected by copyright, with all rights reserved, unless otherwise indicated.


  Last updated: May 2010. Copyright © Addis Ababa University Libraries - Feedback