Information Technology Reference
In-Depth Information
Chapter 7
Case-Based Reasoning in the Health Sciences:
Foundations and Research Directions
Isabelle Bichindaritz
1
and Cindy Marling
2
1
University of Washington, Tacoma, Institute of Technology
1900 Commerce Street,
Tacoma, Washington 98402, USA
ibichind@u.washington.edu
2
School of Electrical Engineering and Computer Science,
Russ College of Engineering and Technology,
Ohio University,
Athens, Ohio 45701, USA
marling@ohio.edu
Abstract.
Case-based reasoning (CBR) is an Artificial Intelligence
(AI) approach with broad applicability to building intelligent systems
in health sciences domains. It represents knowledge in the form of
exemplary past experiences, or cases. It is especially well suited to health
sciences domains, where experience plays a major role in acquiring knowl-
edge and skill, and where case histories inform current practice. This
chapter provides a broad overview of CBR in the Health Sciences, in-
cluding its foundations and research directions. It begins with introduc-
tions to the CBR approach and to health sciences domains, and then
explains their synergistic combination. It continues with a discussion of
the relationship between CBR and statistical data analysis, and shows
how CBR supports evidence-based practice. Next, it presents an in-depth
analysis of current work in the field, classifying CBR in the Health Sci-
ences systems in terms of their domains, purposes, memory and case
management, reasoning, and system design. Finally, it places CBR with
respect to other AI approaches used in health sciences domains, showing
how CBR can complement these approaches in multimodal reasoning
systems.
1
Introduction
Case-based reasoning (CBR) is an Artificial Intelligence (AI) approach with
broad applicability to building intelligent systems in health sciences domains.
Biomedical domains have provided fertile ground for AI research from the earliest
days. Moreover, the healthcare domain is one of the leading industrial domains
in which computer science is applied today. The value of CBR stems from cap-
turing specific clinical experience and leveraging this contextual, instance-based,
knowledge for solving clinical problems. CBR systems have already proven use-
ful in clinical practice for decision support, explanation, and quality control [1].
Search WWH ::
Custom Search