Benefits of embedding structural constraints in coherent diagnostic processes

This paper reviews recent results applicable to medical diagnosis, obtained by adding structural constraints to a coherent inference process. Such further considerations turn out to be useful whenever a basic lower-upper conditional probability assessment induces extension bounds too vague to motivate an informed decision. Three general types of qualitative judgements are proposed and fully described. They do not constitute a “panacea” to solve every problematic situation, but their application can considerably improve inferences results in specific cases, as two practical applications show.