A new expert system for histopathological diagnosis of human ovarian epithelial cancer.

A new rule-based expert system designed to assist pathologists in the histological diagnosis of human ovarian epithelial cancer has been developed. The system was implemented using the INSIGHT II+ EXPERT System Shell and runs on IBM compatible microcomputers. It operates using knowledge acquired through the literature, research and experience, formulated into a total of 195 IF-THEN rules. The WHO International Histological Classification of Tumours is being used throughout. The system can quickly focus on a differential problem, thereby reducing the time necessary to reach a conclusion. It has been tested on a set of 34 cases, previously examined by pathologists of the Metaxas Memorial Cancer Institute, and has been found to reach good agreement with the pathologists' diagnoses.