![]() |
|||
![]()
|
![]() |
![]() |
![]() |
Chapter 32
|
1. | Introduction | |
2. | Historical Account | |
3. | Methodologies | |
3.1. | Bayesian Statistical Approaches | |
3.2. | Rule-Based Reasoning | |
3.3. | Neural Network | |
3.4. | Case-Based Reasoning | |
3.5. | Object-Oriented Programming | |
4. | The Status of Applications Research | |
4.1. | INTERNIST/CADUCEUS: An Expert System in Internal Medicine | |
4.2. | MYCIN: An Expert System in Blood Infections | |
4.3. | PUFF: An Expert System in Pulmonary Disorders | |
4.4. | QMR: A Medical Diagnostic Expert System | |
4.5. | EEG Analysis System | |
4.6. | Other Expert Systems Developed in the 1990s | |
5. | Research Issues | |
5.1. | Selecting a Medical Domain | |
5.2. | Knowledge Acquisition and Representation | |
5.3. | Validation and Verification | |
5.4. | Ethical and Legal Considerations | |
5.5. | Acceptability Issues | |
6. | Future Trends | |
6.1. | Processing of Temporal Knowledge | |
6.2. | System Integration | |
6.3. | System Evaluation | |
References | ||
Appendix: Medical Expert Systems Developed from 1992 to 1996 |
This chapter reviews the research on expert systems designed as aids to medical decision-making. Because complex medical decisions are often made when major uncertainties are present and when the stakes are extremely high, expert systems are ideally suited for decision analysis. Over the last 30 years, many medical expert systems have been developed. Motivations for the development of expert systems in medicine have been numerous. Assisting physicians in making diagnoses and treatment recommendations is the most commonly found application of expert systems in medicine. A physician may have knowledge of most diseases, but, due to the extensive number of diseases, a physician could benefit from the support provided by an expert system to quickly isolate the disease. Specifically, the goals of developing expert systems for medicine are as follows (Shortliffe et al., 1979):
In medical diagnosis, the exhaustive nature of problem-solving in expert systems to ensure that remote possibilities are not overlooked is important (Barr and Feigenbaum, 1982). Often, the very codification of expertise in suitable form for an expert systems is an illuminating and valuable part of the expert systems development. Furthermore, since the nature of the problem in medicine is not sufficiently understood, large amounts of domain-specific knowledge have to be represented and reasoned with.
Medical treatment decisions are another prime area for applying expert systems. This is so because such decisions must typically be made by patients and their physicians at a point in the diagnostic process that has produced a relatively stable determination of the patient's problem and where specific treatment or more invasive testing options are being considered.
An expert system should be used in medical practice only if it improves the quality of care at an acceptable cost in time or money, or if it maintains the existing standard of care at a reduced cost in time or money. Miller et al. (1985) defined improved quality of care by one or more of the following criteria: improved diagnositc accuracy; improved therapeutic results; an improved sense of the patients well-being; easier and more rapid access to patient information via better record-keeping systems; and a better representation of facts in medical records and better documentation of the reasons for the physicians' actions. These criteria may be used in evaluating medical expert systems when they are applied in a clinical setting.
The development of medical expert systems brings with it many formidable technical, behavioral, legal, and ethical problems that must be addressed by the researchers in this field. These include acquiring and representing medical knowledge, validating the systems, getting physicians and patients to accept them, and deciding the responsibility for clinical decisions made with the help of these systems. This chapter reviews historical accounts of expert systems in medicine, key techniques, applications, research issues, and future trends of medical expert systems.
Previous | Table of Contents | Next |
![]() |
|
Use of this site is subject certain Terms & Conditions. Copyright (c) 1996-1999 EarthWeb, Inc.. All rights reserved. Reproduction in whole or in part in any form or medium without express written permission of EarthWeb is prohibited. Please read our privacy policy for details. |