Enhanced parameter identification for complex biomedical models on the basis of fuzzy arithmetic

With the objective of future improvement of medical therapy, a method for the improvement of parameter identification for complex biomedical models on the basis of fuzzy arithmetic is presented. The model parameters to be identified are considered as uncertain, and are represented by fuzzy numbers with their membership functions quantifying an initial guess for the actual value of the model parameter. Using the transformation method as a special implementation of fuzzy arithmetic, the model can be analyzed with the intention of determining the influence of each parameter on the variation of the overall model output. Finally, the parameter identification can be improved by reducing the high-dimensional identification procedure to a number of lower-dimensional optimization problems with measurement data taken from those time intervals only where the parameters show significant influence. The enhanced performance of the model with the newly identified parameters is proved by a higher conformity of its predictions with reality.