An idea of the application of fuzzy neural networks to medical decision support systems

摘要

The paper presents a methodology for medical decision support systems design which combines a fuzzy-set approach with artificial neural networks. The proposed methodology can deal with two types of medical knowledge which usually contribute to medical diagnosis: fuzzy knowledge (linguistic rules provided by human experts); and nonfuzzy knowledge (numerical data obtained from medical tests). The proposed approach has been successfully applied in medical diagnosis as a support system in the treatment of duodenal ulcers with the use of highly selective vagotomy.