Representing Causation in Medicine: How Fuzzy "Sets as Points" Capture the Uniqueness of the Patient, Context, and Change in the Fuzzy Unit Hypercube

Fuzzy logic has found its application in medicine through technology and soft computing. Less studied are clinical implications for medical diagnosis and therapy. In order to lend itself scientifically as a method for "evidence-based medicine" fuzzy logic offers more than probability-based statistics because it can deal with causation and uncertainty. Decisions in certainty assume no role for unknown factors in the analysis. However, in the real patient unknown historical, physiological, environmental, other contextual factors and known but unmeasured variables are always present and affect the clinical course. We show how the fuzzy causation measure K derived from the fuzzy Subsethood theorem characterizes change in patient condition, disambiguates clinical decision and shows how representation of a patient by known variables is but a shadow of the reality of a multitude of unrepresented variables, some of which can be measured, others, which are totally unnamed.