Figure 1

Overview of the complete parameter estimation framework. The estimation process begins by presenting the parameter estimation module with an initial set of kinetic parameters. A first pass through the parameter estimation is performed and used to initialize the identifiability analysis (IA) module. The IA determines and classifies the non-identifiable parameters, suggesting possible solutions. The IA then returns the subset of parameters found to be identifiable in addition to the informed prior which may be used by the CSUKF to formulate the prior state distribution. Thus the CSUKF is able to determine a unique solution, even if some parameters remain non-identifiable.