This research proposes a formal multicenter clinical trial to determine the ability of an expert system with an advanced knowledge base structure and reasoning engine to assist pathologists in making diagnostic decisions about diagnostic problems in lymph node pathology. For over five years, the Pathfinder project has focused on theoretical and pragmatic problems posed by the construction of a computer-based decision system for use in lymph node pathology by general pathologists. Our work has focused on identified problems with the diagnosis of lymph node diseases; but it has also been concerned with solving the fundamental problems associated with uncertain reasoning in artificial intelligence research. In the course of this work, we encountered and solved difficult problems surrounding the acquisition, representation, and manipulation of knowledge about the uncertain relationships among features and diseases in lymph node pathology. The techniques that we have developed have reached maturity. We now plan to carry out a randomized trial of the use of the system with general pathologists. This will be designed to accumulate the necessary data for a statistical analysis of the efficacy of the system in providing assistance with the diagnostic subproblems of feature identification and integration in morphologic diagnosis. At the same time, we will carry out a qualitative study of pathologists' attitudes on the use of computer-based decision support systems in the clinical environment.
| Nathwani, B N; Clarke, K; Lincoln, T et al. (1997) Evaluation of an expert system on lymph node pathology. Hum Pathol 28:1097-110 |
| Heckerman, D E; Horvitz, E J; Nathwani, B N (1992) Toward normative expert systems: Part I. The Pathfinder project. Methods Inf Med 31:90-105 |
| Heckerman, D E; Nathwani, B N (1992) An evaluation of the diagnostic accuracy of Pathfinder. Comput Biomed Res 25:56-74 |
| Heckerman, D E; Nathwani, B N (1992) Toward normative expert systems: Part II. Probability-based representations for efficient knowledge acquisition and inference. Methods Inf Med 31:106-16 |
