2.50
Hdl Handle:
http://hdl.handle.net/10755/154941
Type:
Presentation
Title:
Knowledge acquisition for assessment of preterm labor risk in pregnant women
Abstract:
Knowledge acquisition for assessment of preterm labor risk in pregnant women
Conference Sponsor:Sigma Theta Tau International
Conference Year:1992
Conference Date:August 6 - 8, 1992
Author:Woolery, Linda, DPH/DrPH
P.I. Institution Name:University of Missouri
School of Nursing
Title:Assistant Professor
Problem, Conceptual Base, Literature: The exact causes which initiate preterm labor remain mostly unknown. Early detection and treatment of preterm labor can prolong gestation with improved outcomes for both the infant and family. Review of the literature provided both theoretical and empirical support to generate and test a knowledge base for nurses' assessment of preterm labor risk, where epidemiologic factors, uterine and cervical factors, biophysical factors, infections, biochemical factors, psychosocial factors, prenatal education and other interventions were reported within the knowledge domain for preterm labor risk assessment. Preterm labor risk screening tools reviewed included factors which were not valid predictors of preterm labor risk and failed to include factors reported in the literature, such as hematocrit, infection, biochemical factors, etc., that may be valid predictors of preterm labor risk. And though existing scoring and screening tools are not adequately predictive for preterm labor risk, nor do they meet psychometric standards for reliability and validity, these tools are used, on a daily basis, to intervene with pregnant women.



Numerous authors and agencies have published theoretical papers, but little empirical literature, on the need for research and knowledge development for computer applications to support clinical nursing practice. Knowledge acquisition techniques offer a research and development platform which uses inductive learning techniques with `real' patient data already collected by nurses and stored in computerized databases. This methodology was supported through earlier work by Woolery, Grzymala-Busse, Summers, and Budihardjo (1991), and meets Hinshaw's (1989) challenge to develop a knowledge base for the nursing profession.



Research Questions: What data items are statistically significant predictors of preterm labor risk? What production rules are generated from the data by an ID3 knowledge acquisition software program? What are the verified rules in the knowledge base after content validity procedures are conducted with nurse experts?



Design/Method: A predictive correlational study design was used to statistically analyze data for 2400 randomly selected subjects in a large existing clinical database (instrument) containing data for over 16,000 high risk pregnant women (sample). Both ID3 and expert verification techniques were used for knowledge acquisition as a part of the methodology.



Procedures: After IRB approval, a copy of a large existing perinatal database was obtained. The data were analyzed using SPSS multiple regression techniques. Statistically significant predictor variable sets were used as attribute partitions for ID3 program runs to generate production rules from the data. Next a panel of nurse experts *will conduct content validity procedures to verify production rule output. These procedures *will differentiate verified rules from those that were judged 'flawed' by the experts. All verified rules *will be entered into a knowledge base for future prototype expert system development. Experts will also provide lists of missing data and missing rules needed for further knowledge base development.



Findings: Statistical and ID3 analysis are completed and expert verification techniques are *in progress.. The knowledge base from this research will, in the future, be used for an expert system to provide clinical decision support to perinatal nurses. The knowledge acquisition methodology used seems successful and offers a data-driven inductive approach to knowledge base development for nursing.
Repository Posting Date:
26-Oct-2011
Date of Publication:
6-Aug-1992
Sponsors:
Sigma Theta Tau International

Full metadata record

DC FieldValue Language
dc.typePresentationen_GB
dc.titleKnowledge acquisition for assessment of preterm labor risk in pregnant womenen_GB
dc.identifier.urihttp://hdl.handle.net/10755/154941-
dc.description.abstract<table><tr><td colspan="2" class="item-title">Knowledge acquisition for assessment of preterm labor risk in pregnant women</td></tr><tr class="item-sponsor"><td class="label">Conference Sponsor:</td><td class="value">Sigma Theta Tau International</td></tr><tr class="item-year"><td class="label">Conference Year:</td><td class="value">1992</td></tr><tr class="item-conference-date"><td class="label">Conference Date:</td><td class="value">August 6 - 8, 1992</td></tr><tr class="item-author"><td class="label">Author:</td><td class="value">Woolery, Linda, DPH/DrPH</td></tr><tr class="item-institute"><td class="label">P.I. Institution Name:</td><td class="value">University of Missouri<br/>School of Nursing</td></tr><tr class="item-author-title"><td class="label">Title:</td><td class="value">Assistant Professor</td></tr><tr><td colspan="2" class="item-abstract">Problem, Conceptual Base, Literature: The exact causes which initiate preterm labor remain mostly unknown. Early detection and treatment of preterm labor can prolong gestation with improved outcomes for both the infant and family. Review of the literature provided both theoretical and empirical support to generate and test a knowledge base for nurses' assessment of preterm labor risk, where epidemiologic factors, uterine and cervical factors, biophysical factors, infections, biochemical factors, psychosocial factors, prenatal education and other interventions were reported within the knowledge domain for preterm labor risk assessment. Preterm labor risk screening tools reviewed included factors which were not valid predictors of preterm labor risk and failed to include factors reported in the literature, such as hematocrit, infection, biochemical factors, etc., that may be valid predictors of preterm labor risk. And though existing scoring and screening tools are not adequately predictive for preterm labor risk, nor do they meet psychometric standards for reliability and validity, these tools are used, on a daily basis, to intervene with pregnant women.<br/><br/><br/><br/>Numerous authors and agencies have published theoretical papers, but little empirical literature, on the need for research and knowledge development for computer applications to support clinical nursing practice. Knowledge acquisition techniques offer a research and development platform which uses inductive learning techniques with `real' patient data already collected by nurses and stored in computerized databases. This methodology was supported through earlier work by Woolery, Grzymala-Busse, Summers, and Budihardjo (1991), and meets Hinshaw's (1989) challenge to develop a knowledge base for the nursing profession.<br/><br/><br/><br/>Research Questions: What data items are statistically significant predictors of preterm labor risk? What production rules are generated from the data by an ID3 knowledge acquisition software program? What are the verified rules in the knowledge base after content validity procedures are conducted with nurse experts?<br/><br/><br/><br/>Design/Method: A predictive correlational study design was used to statistically analyze data for 2400 randomly selected subjects in a large existing clinical database (instrument) containing data for over 16,000 high risk pregnant women (sample). Both ID3 and expert verification techniques were used for knowledge acquisition as a part of the methodology.<br/><br/><br/><br/>Procedures: After IRB approval, a copy of a large existing perinatal database was obtained. The data were analyzed using SPSS multiple regression techniques. Statistically significant predictor variable sets were used as attribute partitions for ID3 program runs to generate production rules from the data. Next a panel of nurse experts *will conduct content validity procedures to verify production rule output. These procedures *will differentiate verified rules from those that were judged 'flawed' by the experts. All verified rules *will be entered into a knowledge base for future prototype expert system development. Experts will also provide lists of missing data and missing rules needed for further knowledge base development.<br/><br/><br/><br/>Findings: Statistical and ID3 analysis are completed and expert verification techniques are *in progress.. The knowledge base from this research will, in the future, be used for an expert system to provide clinical decision support to perinatal nurses. The knowledge acquisition methodology used seems successful and offers a data-driven inductive approach to knowledge base development for nursing.</td></tr></table>en_GB
dc.date.available2011-10-26T13:24:10Z-
dc.date.issued1992-08-06en_GB
dc.date.accessioned2011-10-26T13:24:10Z-
dc.description.sponsorshipSigma Theta Tau Internationalen_GB
All Items in this repository are protected by copyright, with all rights reserved, unless otherwise indicated.