Biomedical knowledge acquisition : three systems reviewed

Technical Report

Analytics

Citeable URL: https://ir.library.oregonstate.edu/concern/technical_reports/k930c5462

Attribute Name	Values
Alternative Title	86-60-1
Creator	Springsteel, Frederick Neil
Abstract	Recent advances in diagnostic expert systems have been limited by the crucial Knowledge Acquisition Problem: to bring expert heuristics into logical, conditional rule forms useful for computer aided diagnosis. Some automatic systems with potential biomedical relevance that do acquire knowledge are reviewed here, and analyzed from the standpoint of possible scientific advancement. The reviewed projects are: PLANT/ds, for diagnosis of crops diseases; RX, for study of rheumatic cases and treatments; and the statistics oriented GUHA-80 strategy, which has been applied medically but has more general aims. The last two projects are of more interest to clinical and biometrical researchers, respectively, while the first one has been more clearly successful in automating diagnostic knowledge acquisition. Comparisons of their three different learning strategies then leads to preliminary suggestions for combining their positive features into a second-level, enhanced knowledge acquisition system (EKAS).
Resource Type	Technical Report
Date Issued	1986
Academic Affiliation	Computer Science
Series	Technical report (Oregon State University. Department of Computer Science)
Rights Statement	In Copyright
Publisher	Oregon State University. Department of Computer Science
Language	English [eng]
File Format	pdf

Thumbnail	Title	Date Uploaded	Visibility	Actions
	Springsteel_Frederick_Neil_86_60_01.pdf	2023-01-31	Public	Download