Abstract
This paper proposes a Clinical Reasoning System based on clinicians' diagnostic model to support medical diagnosis. The paper first clarifies the model of clinician's diagnosis process and then discusses about how the clinician reason their diagnosis. In the clinical field, it is necessary to detect important deceases at early stage of medical diagnosis without considering their occurrence frequency or possibility. Therefore, the proposed system applies inverse posibilistic causal reasoning using uncertain disease-symptom knowledge base to conclude the disease from the measured symptoms. Most of the proposed diagnosis systems only support single-decease assumptions. However, based on underlying cause-effect relations, when resulting causal model to diagnostic problems, multiple deceases may occur simultaneously This paper also discusses the method for multiple-decease assumptions in clinical reasoning.
