Membership, syndromic and causality graphs to represent the emergency physician’s reasoning

Abstract

The reasoning of an emergency doctor is very particular because he often has no prior knowledge (he generally knows nothing about the patient), he has little time to understand the situation, and he must, by means of questioning, quickly make a decision based on diagnostic hypotheses and an estimate of the seriousness of the patient’s condition. In a previous paper we presented the 10 step reasoning model of an emergency physician used by the symptom checker MedVIR. This model is represented by a membership graph which describes the clinical information that helps physicians in their diagnosis by allowing them to ask the essential questions.This paper extends membership graphs to syndromic vision as well as to causality graphs. Syndromic graphs enable the simplification of the memorization of a disease whereas causality graphs make it possible to define the first cause of a disease, the chain of pathological processes, the syndromes which constitute it. These graphs enable the clinician to have a global view of diseases belonging to different and unrelated specialities and they will be able to contribute to the training of doctors and health professionals.