A FUZZY KNOWLEDGE-BASED SYSTEM FOR BIOMEDICAL IMAGE INTERPRETATION

A general purpose knowledge-based system for biomedical image interpretation is presented. The system acquires knowledge directly from the experts by means of a user friendly dialogue. The knowledge introduced tailors the system to a particular biomedical application. Frame representation technique is used for the representation of descriptive knowledge and a fuzzy reasoning strategy, based on fuzzy production rules, is adopted to manipulate the certain and uncertain knowledge contained into Frame Slots and to deduce interpretations. A detailed description of the application of the system to the analysis of CT images of vertebrae for the quantity evaluation of the bone mineral content is provided.