Knowledge Representation and Cases indexing in Upper Digestive Endoscopy
Résumé
Provided with evolved functionalities, a digestive endoscopy atlas can be used as a training tool and even as a diagnostic aid for "junior" endoscopists. Next to a classic illustration of diagnoses by means of noteworthy iconography, such an atlas must in addition integrate similar cases retrieval and diagnostic hypothesis evaluation. That is precisely the main reason why a Case Based Reasoning (CBR), organized around two bases, one of endoscopic knowledge and one of case iconography, is well adapted to the objectives; actually, the CBR cycle-new case description, similar erases retrieval, knowledge base correction and new case training-allows one to conceive a system able to improve. This paper focuses on the creation of two bases, in other words, the knowledge representation and the iconography indexing. The concept of Scenes with Objects is useful to describe the endoscopic diagnoses as well as the images or sequences of the cases. The endoscopic diagnosis description obviously falls within the province of an expert. He/she generally prefers to express uncertainty and vagueness in a qualitative form rather than in a quantitative one. So, merging Rough Set Theory and Qualitative Theory of Uncertainty appears to be an unexplored but promising way in order to elaborate a Knowledge Based System. Moreover, thanks to description common mode, the knowledge base takes the place of thesaurus valuable to the cases indexing. Likewise, the elaborated knowledge representation and indexing open some tracks towards the similar ease retrieval.