Optimal experimental design with the sigma point method.

Max-Planck-Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany.
IET Systems Biology (Impact Factor: 1.67). 02/2009; 3(1):10-23. DOI: 10.1049/iet-syb:20080094
Source: PubMed

ABSTRACT Using mathematical models for a quantitative description of dynamical systems requires the identification of uncertain parameters by minimising the difference between simulation and measurement. Owing to the measurement noise also, the estimated parameters possess an uncertainty expressed by their variances. To obtain highly predictive models, very precise parameters are needed. The optimal experimental design (OED) as a numerical optimisation method is used to reduce the parameter uncertainty by minimising the parameter variances iteratively. A frequently applied method to define a cost function for OED is based on the inverse of the Fisher information matrix. The application of this traditional method has at least two shortcomings for models that are nonlinear in their parameters: (i) it gives only a lower bound of the parameter variances and (ii) the bias of the estimator is neglected. Here, the authors show that by applying the sigma point (SP) method a better approximation of characteristic values of the parameter statistics can be obtained, which has a direct benefit on OED. An additional advantage of the SP method is that it can also be used to investigate the influence of the parameter uncertainties on the simulation results. The SP method is demonstrated for the example of a widely used biological model.


Available from: Andreas Kremling, Feb 28, 2014
  • Source
    11th Symposium on Process Systems Engineering, Singapore, (Jul 15-17, 2012); 07/2012
  • [Show abstract] [Hide abstract]
    ABSTRACT: Informative experiments are highly valuable for estimating parameters in nonlinear dynamic bioprocesses. Techniques for optimal experiment design ensure the systematic design of such informative experiments. The E-criterion which can be used as objective function in optimal experiment design requires the maximization of the smallest eigenvalue of the Fisher information matrix. However, one problem with the minimal eigenvalue function is that it can be nondifferentiable. In addition, no closed form expression exists for the computation of eigenvalues of a matrix larger than a 4 by 4 one. As eigenvalues are normally computed with iterative methods, state-of-the-art optimal control solvers are not able to exploit automatic differentiation to compute the derivatives with respect to the decision variables. In the current paper a reformulation strategy from the field of convex optimization is suggested to circumvent these difficulties. This reformulation requires the inclusion of a matrix inequality constraint involving positive semidefiniteness. In this paper, this positive semidefiniteness constraint is imposed via Sylverster's criterion. As a result the maximization of the minimum eigenvalue function can be formulated in standard optimal control solvers through the addition of nonlinear constraints. The presented methodology is successfully illustrated with a case study from the field of predictive microbiology. Copyright © 2015. Published by Elsevier Inc.
    Mathematical Biosciences 02/2015; 264. DOI:10.1016/j.mbs.2015.02.006 · 1.49 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The identification of a model structure, i.e., the relationship of model components, and the usually unknown model parameters are essential to get meaningful deterministic models. Due to imperfect and limited measurement data to fulfil both requirements can be a challenging task. Therefore, the Optimal Experimental Design (OED) was introduced to detect some optimal stimuli of the regarded system, that increase the information content of measurements. For models which are linear with respect to their parameters, the OED is well known to reduce the influence of measurement noise to the parameter estimation process and to find conditions that make the difference of competing model hypothesis more obvious. Unfortunately, in practise the most used models are non-linear with respect to their parameters. The established methods for the linear case, e.g. the Fisher Information matrix (FIM), can lead to a crude approximation of the non-linearity that results at best in a sub-optimal choice of the system stimuli. The usage of sample based approaches, e.g. Monte Carlo methods, seems to be more suitable to consider non-linear effects, but their inherent computational effort prohibits the application in the framework of Optimal Experimental Design. In this work, the Sigma Point method is used instead, because its advantages are at least threefold for OED: 1) a more reliable approximation of parameter uncertainties can be achieved, 2) a definition of a novel design criterion for OED becomes possible, and 3) it provides an efficient and elegant way to take the imperfection of parameter estimation in the field of model selection into account. All three aspects are demonstrated for the example of a biological substrate uptake model.
    Mathmod 2009 - 6th Vienna International Conference on Mathematical Modelling; 01/2009