Transcript Representing Nursing Knowledge Applications for Database

Representing Nursing Knowledge
Applications for Database Design
Josette Jones, RNc
Patricia Brennan, RN, PhD
Presentation Overview






Introduction
Background and Significance
Role of Knowledge Representation Systems
Indexing WebPages Using MESH
Information Retrieval
Evaluation
University of Wisconsin–Eau Claire
Introduction

Effective Use of Information in HealthCare
–
–

Accessible context
Matching to individual needs
Indexing and Organizing On-line Resources
–
–
Content
Anticipated Usage
University of Wisconsin–Eau Claire
Background and Significance
Changes in healthcare impact
discharge teaching
Patient-specific health information
is available on the WWW
University of Wisconsin–Eau Claire
Role of
Knowledge Representations
Systems in Indexing

Indexing techniques
Description attributes
Keyword attributes
•
Knowledge Representation Systems (KRS)
University of Wisconsin–Eau Claire
Knowledge Representation
Systems

A Knowledge Representation System has:
–
–
–

An underlying knowledge representation
language (meta-language) with its vocabulary and
explicit structure
A semantic (meaning of the expressions of the
language)
A restricted syntax (set of reasoning rules)
Examples of Health Care / Nursing Knowledge
Representation Systems
University of Wisconsin–Eau Claire
Indexing WebPages Using a
Medical Thesaurus
University of Wisconsin–Eau Claire
The HeartCare Project

Providing health information
–
–

Graduated to patient’s stage of recovery
Tailored to his/her medical profile and individual needs
Filtered set of cardiac recovery resources available
on the web stored in an Access© database
–
–
–
Self constructed web pages
Web pages are described with index terms
Index terms describe the medical profile
Matching algorithm web page – patient
University of Wisconsin–Eau Claire
Indexing Web Pages
in HeartCare
Nurse-clinicians tagged web documents with:
Selected concepts from Medical
Language Subject Heading (MeSH)
Supplemented with terms reflecting
local clinical practice
University of Wisconsin–Eau Claire
Example of Indexing
http://www.women.americanheart.org/physicians/sub_content/ten.html
tagged with the terms “diet” and “weight” is pulled 4 different
times for the menu heading “Ten questions a woman should ask her
healthcare provider”
http://www.amhrt.org/Heart_and_Stroke_A_Z_Guide/calccb.html
tagged with terms “Beta Blockers/Calcium Channel Blockers”and
“Medications” are pulled for all conditions that have the subject
heading assigned, even when not applicable
University of Wisconsin–Eau Claire
Implications for Retrieval
Too many pages pulled per patient
Too many duplicate pages
Some pages were pulled that did not
exactly match the patient profile
University of Wisconsin–Eau Claire
Examples of Total Web Pages in
Combination with Menu Title
Retrieved for Patients
Patient
Combinations
Retrieved
Unique
Combinations
Patient 1
266
138
Patient 2
891
647
Patient 3
324
281
Patient 4
584
203
Example of Duplicate Page Retrieval
Using Keywords
“smoking and behavior changes”
Menu Title
Condition
Taking charge of your health - Week 3-6
Diabetes
Taking charge of your health - Week 3-6
Hypertension
Taking charge of your health - Week 3-6
Smoking
Beginning lifestyle changes - Week 7-12
Hypertension
Beginning lifestyle changes - Week 7-12
Smoking
Beginning lifestyle changes - Week 7-12
Diabetes
Changing your lifestyle - Week 13-26
Diabetes
Changing your lifestyle - Week 13-26
Hypertension
Changing your lifestyle - Week 13-26
Smoking
http://rex.nci.nih.gov/NCI_Pub_Interface/Clearing_the_Air/clearing.html
Pages Retrieved that Does not
Match the Patient’s Profile
Sample Male Patient with Risk
Factors Hypertension and Stress
# of Web pages
Non-matching topic
11
Risk of smoking and smoking cessation
8
Risk factors for women
5
Being overweight and weight loss
2
Diabetes management
University of Wisconsin–Eau Claire
Evaluation
Flawed indexing system
Lacking structure of index terms
Conceptualization problem
University of Wisconsin–Eau Claire
Discussion
Keywords must be part of semantic
representation understood by users and indexers
Relation content and usage must be explicated
Keywords must converge
University of Wisconsin–Eau Claire
This study is supported by NLM/NINR Grant
LM06249, Principal Investigator Dr. P.F. Brennan
The authors want to thank the members from the
HeartCare team for their advice and support.
University of Wisconsin–Eau Claire
Josette Jones
Patricia F. Brennan
[email protected]
University of Wisconsin
Eau Claire
[email protected]
University of Wisconsin
Madison