Fuzzy Relationships and Fuzzy Control in Medical Knowledge-Based Systems

Some of the first knowledge-based systems to be introduced were medical knowledge-based systems, namely MYCIN, INTERNIST, CASNET, PIP, EXPERT, and CADIAG. The latter is one of the first to use the theory of fuzzy sets. It was developed to assist the physician in diagnostics. The present article delineates two specific pathways resulting from a bifurcation in the history of applied fuzzy systems in medicine. This bifurcation occurred in the 1970’s in the history of the theory of fuzzy sets and systems, when Lotfi A. Zadeh published the “rule of max-min composition” and other researchers applied this rule in different areas. This was the origin of two research areas: fuzzy control, initiated by Sedrak Assilian and Ebrahim Mamdani in London, and fuzzy relations, introduced by Elie Sanchez in Marseille. Later on both concepts were used to construct medical knowledge-based systems in medicine. We present two Viennese systems representing these concepts: the “fuzzy version” of the Computer-Assisted DIAGnostic System (CADIAG) which was developed at the end of the 1970s, and a fuzzy knowledge-based control system, FuzzyKBWean, which was established as a real-time application based on the use of a Patient Data Management System (PDMS) in the intensive care unit (ICU )in 1996.