Computer-based expert systems are fast becoming an indispensable part of modern medicine. These systems are now being widely implemented to solve clinical problems and to evaluate and test students and practitioners. Providing high-quality knowledge bases to support the function of these systems is a top priority. However, modeling the decision strategies of human medical experts for such knowledge bases is both dffficult and expensive. We have substantial experience with these problems, as we have created and maintained several important knowledge bases. During this time, we have worked on developing new tools to support and improve the knowledge engineering work. Several of these prototype tools are now almost ready to be applied outside of the laboratory in diverse, everyday knowledge engineering situations. This application proposes to perform the final developmental work required to implement these tools in realistic settings. The application also proposes to experimentally evaluate the tools in these settings. We expect our experimental work will show a significant improvement in the diagnostic reliability, accuracy, and validity of frames when the tools are applied. In addition, we expect to show that the tools can economize on knowledge engineering effort, thereby reducing the cost of the knowledge engineering work. These tools will provide practical solutions to the knowedge engineering problems we have identified.
| Lincoln, M J; Turner, C W; Haug, P J et al. (1992) Iliad's role in the generalization of learning across a medical domain. Proc Annu Symp Comput Appl Med Care :174-8 |
| Lincoln, M J; Turner, C W; Haug, P J et al. (1991) Iliad training enhances medical students' diagnostic skills. J Med Syst 15:93-110 |
