MEDICAL
INFORMATICS
Knowledge Management
and Data Mining in
Biomedicine
INTEGRATED SERIES IN INFORMATION SYSTEMS
Series Editors
Professor Ramesh Sharda Prof. Dr. Stefan Vo13
Oklahoma State University Universitat Hamburg
Other published titles in the series:
E-BUSINESS MANAGEMENT:
Integration of Web Technologies with Business
Models1
Michael
J.
Shaw
VIRTUAL CORPORATE UNIVERSITIES:
A Matrix of Knowledge and Learning
for the New Digital
DawdWalter R.J. Baets
&
Gert Van der Linden
SCALABLE ENTERPRISE SYSTEMS:
An Introduction to Recent Advances1
edited by Vittal Prabhu, Soundar Kumara, Manjunath Kamath
LEGAL PROGRAMMING:
Legal Compliance for RFID and Software Agent
Ecosystems in Retail Processes and Beyond1
Brian Subirana and Malcolm Bain
LOGICAL DATA MODELING:
What It Is and How To Do It1

Carol Friedman William Hersh
Columbia University, USA Oregon Health
&
Science Univ., USA
Library of Congress Cataloging-in-Publication Data
A
C.I.P. Catalogue record for this book is available
from the Library of Congress.
ISBN-10: 0-387-2438 1-X (HB)
ISBN- 10: 0-387-25739-X (e-book)
ISBN- 13: 978-0387-2438 1-8 (HB) ISBN- 13: 978-0387-25739-6 (e-book)
O
2005 by Springer Science+Business Media, Inc.
All rights reserved. This work may not be translated or copied in whole or in
part without the written permission of the publisher (Springer Science
+
Business Media, Inc., 233 Spring Street, New York, NY 10013, USA), except
for brief excerpts in connection with reviews or scholarly analysis. Use in
connection with any form of information storage and retrieval, electronic
adaptation, computer software, or by similar or dissimilar methodology now
know or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks and
similar terms, even if the are not identified as such, is not to be taken as an
expression of opinion as to whether or not they are subject to proprietary rights.
Printed in the United States of America.
98765432
1
SPIN
1
1055556

Knowledge Management, Data Mining, and Text Mining
Applications in Biomedicine

12
3.1 Ontologies

13
3.2 Knowledge Management

14
3.3 Data Mining and Text Mining

18
3.4 Ethical and Legal Issues for Data Mining

22
Summary

22
References

23
Suggested Readings

31
Online Resources
31

41
4
.
Data Description

42
5
.
Results

44
5.1 Basic Analysis

44
5.2 Content Map Analysis

47
5.3 Citation Network Analysis

55

6
.
Conclusion and Discussion 57

7
.
Acknowledgement 58

References

Biological Perspective

79

3
.
Case Study 83
3.1
Informatics Perspective
-
The BIOINFOMED Study
and Genomic Medicine

83
3.2 Biological Perspective
-
The BioResearch Liaison

Program at the University of Washington
85
4
.
Conclusions and Discussion

89

5
.
Acknowledgements 91


99
3
.
Review of the Literature: Data Mining and Privacy

and Security 109
vii
3.1
General Approaches to Assuring Appropriate Use

110
3.2 Specific Approaches to Achieving Data Anonymity

112
3.3 Other Issues in Emerging "Privacy Technology"

116
3.4 "Value Sensitive Design": A Synthetic
Approach to Technological Development

117
3.5 Responsibility of Medical Investigators

119
4
.
Case Study: The Terrorist Information Awareness
Program (TIA)

12 1

Online Resources

135
Questions for Discussion

13 7
Chapter
5:
Ethical and Social Challenges of Electronic
Health Information

139
1
.
Introduction

141
2
.
Overview of the Field

142
2.1 Electronic Health Records

142
2.2 Clinical Alerts and Decision Support

146
2.3 Intemet-based Consumer Health Information


Introduction 165

1.1 Use-cases 165
2
.
Context

168
2.1 Concept Characteristics

169
2.2 Domains

170

2.3 Structure 171
3
.
Biomedical Concept Collections

172

3.1 Ontologies 172

3.2 Vocabularies and Terminologies 174

3.3 Aggregation and Classification 175

3.4 Thesauri and Mappings 176
4

Background and Overview: The Use of Concept

Relationships for Knowledge Creation 188

2.1 Indexing Strategies and Vocabulary Systems 190

2.2 Integrating Document Structure in Systems 192
2.3 Text Mining Approaches

194

2.4 Literature-based Discovery IR Systems 195

2.5 Summary 198

.
3 Case Examples 198

3.1 Genescene 199

3.2 Telemakus 200
3.3 How Can a Concept Relationship System Help
with the Researcher's Problem and Questions?

202
3.4 Summary

206
4
.

Ontologies

215
2.1 OpenCyc

215
2.2 WordNet

215
3
.
Examples of Medical Ontologies

217
3.1 GALEN

217
3.2 Unified Medical Language System

219
3.3 The Systematized Nomenclature of Medicine

220
3.4 Foundational Model of Anatomy

222
3.5 MENELAS ontology

223
4

Acknowledgments

232
References

232
Suggested Readings

234
Online Resources

234
Questions for Discussion

235

Appendix: Table showing characteristics of selected ontologies 235

Chapter
9:
Information Retrieval and Digital Libraries
237
Overview of Fields

239
Information Retrieval

241
2.1 Content


4.2 User-oriented Evaluation 265
4.3
Changes in Publishing

267
Acknowledgements

269
References
269
Suggested Readings

273
Online Resources

274
Questions for Discussion

275

Chapter
10:
Modeling Text Retrieval in Biomedicine
277
1
.
Introduction


289
5.2 XplorMed

290
5.3 AI3View:HivResist

291
5.4 The Future

292
xi
References

294
Suggested Readings

295

Online Resources 296
Questions for Discussion

296

Chapter 11: Public Access to Anatomic Images
299
Introduction

301
Background


328
Summary

330
Acknowledgements

330
References

330
Suggested Readings

331
Online Resources

332
Questions for Discussion

332
Chapter 12: 3D Medical Informatics: Information Science
in Multiple Dimensions

333
Introduction

335
Overview
.
3D Medical Informatics

348

4.1 Background and Related work 348
4.2 Design and Software Tools for Template Planning
Workstation

349
4.3 Results and Discussion

350
5
.
Grand Challenges in
3D
Medical Informatics

353
6
.
Conclusion

354
7
.
Acknowledgements

355
References

355

2.4 Infectious Disease Data Analysis and Outbreak
Detection

372
Infectious Disease Information Infrastructure and Outbreak
Detection: Case Studies

378
3.1 New York State's Health Information Network System

378
3.2 The BioPortal System

379

3.3 West Nile Virus Outbreak Analysis 386
Conclusions and Discussion

388
Acknowledgements

391
References


401
2.1 Overview

401
2.2 Levels of Linguistic Structure

402
Domain Knowledge: The UMLS

403
3.1 SPECIALIST Lexicon

404
3.2 Metathesaurus

404
Semantic Network

405
Semantic Interpretation for the Biomedical Literature

406
4.1 Overview

406
4.2 AQUA

407
4.3 PROTEUS-BIO


422

Chapter
15:
Semantic Text Parsing for Patient Records
423
1
.
Introduction
425
2
.
Overview

427
2.1
Challenges of Processing Clinical Reports

427
2.2
Components of an NLP System

431
2.3 Clinical Applications

437
3


from Text Documents 449
Introduction

451
Overview of the Field

453
2.1 Background

453

2.2 Biological Information Extraction 453
2.3 Bioinformatics Tools

456
Case Studies

457

3.1 Identification of Flat Relationships from Text Documents 457
3.2
TransMiner: Formulating Novel, Implicit Associations

through Transitive Closure 461

3.3 Identification of Directional and Hierarchical Relationships 466
BioMap: A Knowledge Base of Biological Literature

477


491
1
.
Introduction
493
2
.
Overview

494
2.1 Metabolic Pathway Databases

494
2.2 Network Modeling and Reconstruction

494
2.3
Extracting Biological Interactions from Text

495
3
.
Metnet

498
3.1

Acknowledgements

514
References

515
Suggested Readings

7 Online Resources

7
Questions for Discussion

518

Chapter 18: Gene Pathway Text Mining and Visualization
519
1
.
Introduction

521
2
.
Literature Review/Overview

S21

542
Suggested Readings

544
Online Resources

545

Questions for Discussion
545

Chapter 19: The Genomic Data Mine
547
1
.
Introduction

549
2
.
Overview

550
2.1 Genomic Text Data

551
2.2 Genomic Map Data

556
xvi

Exploratory Genomic Data Analysis
573
1
.
Introduction

575
2
.
Overview

576
2.1 Gene Expression Data

576

2.2 Mixed Populations 577
2.3 Methods for Mixed Populations

579

2.4 Distance 582
2.5 Hypothesis Selection

584
3
.
Case Studies

586

595
2
.
Overview of the Field

597
2.1 Large-scale Biological Data and Knowledge Resources

597
2.2 Joint Learning Using Multiple Types of Data

599
2.3 Joint Learning Using Data and Knowledge

602
3
.
Kernel-based Data Fusion of Multiple Types of Data

604
3.1 Protein Function Prediction

604
3.2 Kernel-based Protein Function Prediction

604
4
.
Learning Regulatory Networks Using Microarray and
Existing Knowledge

Questions for Discussion

624
Author Index

625
Subject Index

627
EDITORS' BIOGRAPHIES
Hsinchun
Chen
is the McClelland Professor of
Management Information Systems (MIS) at the Eller
College of the University of Arizona. He received his
Ph.D. degree in Information Systems from New York
University. He is the author of more than nine books
and 200 articles covering medical informatics,
knowledge management, homeland security, semantic
retrieval, and Web computing in leading information
technology publications. He serves on the editorial
boards of Journal of the American Society for Information Science and
Technology, ACM Transactions on Information Systems, IEEE Transactions
on Systems, Man, and Cybernetics, IEEE Transactions on Intelligent
Transportation Systems, and Decision Support Systems. He is a scientific
counselor/advisor of the Lister Hill Center of the National Library of
Medicine (NLM/USA) and the National Library of China. Dr. Chen is the
director of the University of Arizona's Artificial Intelligence Lab (40+
researchers). Since 1990, Dr. Chen has received more than $17M in research
funding from various government agencies and major corporations. He has

Department of Medical Education and Biomedical
of Medicine; Professor, Information School and
Adjunct Professor, Department of Health Services, School of Public Health
and Community Medicine. Dr. Fuller has a BA degree in Biology, a Master's
in Library Science from Indiana University, and a Ph.D. in Library and
Information Science from the University of Southern California. Dr. Fuller's
areas of research include: developing new approaches to represent and map
the results
of scientific research; design and evaluation of information
systems to support decision making at the place and time of need; and
integrated health sciences information systems design.
Dr. Fuller serves as Principal Investigator ofthe Health Sciences
Libraries and Information Center contract from the NLM to serve as the
Regional Medical Library for the Pacific Northwest (Alaska, Idaho,
Montana, Washington and Oregon);
Principal Investigator, Telemakus:
Mining and Mapping Research Findings to Promote Knowledge Discovery
in Aging funded by the Ellison Medical Foundation; Co-Investigator of
Biomedical Applications of the Next Generation Internet (NGI): Patient-
centric Tools for Regional Collaborative Cancer Care Using the NGI funded
by the National Library of Medicine; Co-Investigator of an International
Health and Biomedical Research and Training grant from the Fogarty
International Center; and advisor to a Health Services Research
Administration (HRSA) grant to explore models of Faculty Leadership in
Interprofessional Education to Promote Patient Safety.
Dr. Fuller has served as a member of the President's (White House)
Information Technology Advisory Committee and the Board of Regents of
the National Library of Medicine and on the Boards of the American
Medical Informatics Association and the Medical Library Association. She
is an elected fellow of the American College of Medical Informatics. (Email:

York. Dr. Friedman is involved in advancing biomedical informatics, text
mining, and knowledge management research and education. Dr. Friedman
was a conference co-chair for the Natural Language Track of the 2002
Pacific Symposium in Bioinformatics, the Workshop in Biomedicine in the
2002 and 2003 Association for Computational Linguistics Conferences, and
the 2004 BioLink Workshop in the Human Language Techology Conference
of the North American Chapter of the Association for Computational
Linguistics. Dr. Friedman is a member of the Board of Scientific Counselors
of the National Library of Medicine, is a member-at-large of the Executive
Board of the American College of Medical Informatics, is on the Editorial
Boards of the Journal of Biomedical Informatics and the Journal of the
Association of Medical Informatics, and is a reviewer for numerous journals
associated with bioinformatics. Dr. Friedman has been a guest editor of
special issues of the Journal of Biomedical Informatics, has published over
100 articles on NLP, has co-authored a book on natural language processing
(NLP), and is the author of various chapters on NLP.
(Email:
;
URL:

xxii
William
Hersh,
M.D. William Hersh, M.D. is
Professor and Chair of the Department of Medical
Informatics
&
Clinical Epidemiology in the School of
Medicine at Oregon Health
&

Editorial Board of five scientific journals. He is also a member of the
program committee of the Text Retrieval Conference (TREC) and currently
chairs TREC's Genomics Track. Dr. Hersh also serves as Associate Director
of the OHSU Evidence-Based Practice Center funded by the Agency for
Healthcare Research and Quality. Dr. Hersh's work in medical informatics
education is equally well-known. He serves as Director of OHSUYs
educational programs in biomedical informatics. He also teaches medical
informatics to medical students, nursing students, and internal medicine
residents. (Email: ;
URL:

AUTHORS' BIOGRAPHIES
Daniel Berleant, PhD, received the B.S. degree in
1982. After practicing in the software engineering field,
he received the MS (1990) and PhD (1991) degrees from
the University of Texas at Austin. He then developed a
research program in text mining and interaction and in
1
inference under severe uncertainty. In 1999 he accepted a
'I)
position at Iowa State University where he continues to
/
pursue research on text mining and text interaction, as
well as uncertainty quantification and software
engineering. He has advised or co-advised
24
master's theses and six PhD
students who have either graduated or are in progress. He has authored over
50 refereed papers and book chapters. (Email:
;

Hong Kong. He received his PhD degree in
Management Information Systems from the University
of Arizona and a Bachelor degree in Computer Science
!
(Information Systems) from the university of Hong
Kong. He was an active researcher in the Artificial
Intelligence Lab at the University of Arizona, where he
participated in several research projects funded by NSF,
NIH, NIJ, and DARPA. (Email:
;
URL:

Christopher
G.
Chute, MD, DrPH, received his
undergraduate and medical training at Brown University,
internal medicine residency at Dartmouth, and doctoral
training in Epiden~iology at Harvard. He is Board
Certified in Internal Medicine, and a Fellow of the
American College of Physicians, the American College
of Epidemiology, and the American College of Medical
Informatics. He became Head of the Section of Medical
Information Resources at Mayo Foundation in 1988 and
is now Professor and Chair of Biomedical Informatics. As a career scientist
at Mayo, Dr. Chute's NIH and AHCPRIAHRQ funded research in medical
concept representation, clinical information retrieval, and patient data
repositories have been widely published. He is Vice-chair of the ANSI
Health Information Standards Board, Convener of Healthcare Concept
Representation WG3 within the IS0 Health Informatics Technical
Committee, chair-elect of the US delegation to IS0 TC215 for Health


Ted Cooper,
MD, Clinical Associate Professor,
Department of Ophthalmology, Stanford University,
received his MD and completed a residency in
ophthaln~ology at the George Washington University.
He is a fellow of the American College of Medical
Informatics and the American Academy of
Ophthalmology. As National Director for
Confidentiality and Security he helped guide Kaiser
Permanente's response to HIPAA. He has lectured
widely on information assurance. He has participated in a number of health
informatics activities including director and chairperson of the Computer-
based Patient Record Institute. He is currently the chairperson of the Health
Information and Systems Society Privacy and Security Steering Committee,
a member of the Health Information and Systems Society Electronic Health
Record Steering Committee, and chairperson of
The CPRZ Toolkit:
Managing Information Security in Health Care
Work Group. (Email:

;
URL:

pl
Julie Dickerson,
PhD, received her B.S. degree
from the University of California, San Diego and her
MS and PhD degrees from the University of Southern
California. She is currently an Associate Professor of

University. His research interests include systems biology, genetic network
modeling and inference, microarray data analysis, signal processing and
pattern recognition. (Email: )
Shauna Eggers is a computer programmer at the
University of Arizona Artificial Intelligence Lab. She
earned a B.S. in Computer Science and a B.A. in
Linguistics and German Studies from the University of
Arizona in May 2004. Her research interests include
natural language processing for biomedical applications
and knowledge visualization.
Millicent Eidson, MA, DVM, DACVPM
(Epidemiology) is State Public Health Veterinarian and
Director of the Zoonoses Program, New York State
Department of Health. She is also an Associate
Professor in the Department of Epiden~iology,
University at Albany School of Public Health.
Dr.
Eidson previously served as an Epidemic Intelligence
Z
Service (EIS) Officer with the Centers for Disease
Control and Prevention based at the National Cancer