Fuzzy logic and cellular automata allow individual differences in clinical response to be measured

When dosing medication, such as warfarin, serial measures of efficacy 'goal INR' guide the clinician who must also consider the context of the patient in continued dosage. The decision is patient specific and cannot be extrapolated from group data based on probabilities. It is valuable to know the degree to which the warfarin caused the change in measured INR at a given time because this helps the clinician decide the next dosage. We developed the measure K to give the clinician a quantitative estimate of how much a given dose of warfarin causes the change in next measured INR value; K is inversely proportional to the degree [0, 1] that a given dose of coumadin caused the observed effect. Applying cellular automata to each stepped change in patient state, K used as the 'multiplicative factor' and 1/K as the 'initial seed', one can show how K is uniquely attached to the historical contextual time space of that patient and has its own cellular automata pattern.