SPRUS : a knowledge based understanding system for radiology text

Update Item Information
Publication Type dissertation
School or College School of Medicine
Department Biomedical Informatics
Author Ranum, David Leland
Title SPRUS : a knowledge based understanding system for radiology text
Date 1989-08
Description A computerized data acquisition tool, SPRUS (Special Purpose Radiology Understanding System), has been implemented as a module in the HELP Hospital Information System. The tool uses semantic information from a diagnostic expert system to parse free text radiology reports and extract coded findings that are stored in the patient data base. A unique feature of SPRUS is the ability to recognize ambiguous (missing) information and perform resolution based upon the context of the report and the underlying knowledge base. Initial test have shown overall True Positive Rates ranging from 0.86 to 0.90 with Bad Data Rates from 0.046 to 0.06. For diagnostic results, SPRUS shows True Positive Rates from .947 to .986 and Bad Data Rates from 0.04 to 0.06. Errors (both missed and bad) have been categorized and are discussed with examples in terms of management and future project considerations. In summary, SPRUS has prototype of understand knowledge based understanding of radiology reports within the HELP Radiology Subsystem.
Type Text
Publisher University of Utah
Subject Natural Language Processing; Radiology Information Systems; Software Design
Subject MESH Artificial Intelligence; Databases, Factual; Expert Systems; Hospital Information Systems; Medical Informatics Applications
Dissertation Institution University of Utah
Dissertation Name PhD
Language eng
Relation is Version of Digital reproduction of "SPRUS : a knowledge based understanding system for radiology text." Spencer S. Eccles Health Sciences Library.
Rights Management © David Leland Ranum.
Format Medium application/pdf
Format Extent 2,459,182 bytes
Identifier undthes,3801
Source Original: University of Utah Spencer S. Eccles Health Sciences Library (no longer available)
Master File Extent 2,459,306 bytes
ARK ark:/87278/s6vd71cd
Setname ir_etd
ID 191800
Reference URL https://collections.lib.utah.edu/ark:/87278/s6vd71cd