Abstract
Pharmacokinetic/pharmakodynamic (PK/PD) modeling for a single subject is most often performed using nonlinear models based on deterministic ordinary differential equations (ODEs), and the variation between subjects in a population of subjects is described using a population (mixed effects) setup that describes the variation between subjects. The ODE setup implies that the variation for a single subject is described by a single parameter (or vector), namely the variance (covariance) of the residuals. Furthermore the prediction of the states is given as the solution to the ODEs and hence assumed deterministic and can predict the future perfectly. A more realistic approach would be to allow for randomness in the model due to e.g., the model be too simple or errors in input. We describe a modeling and prediction setup which better reflects reality and suggests stochastic differential equations (SDEs) for modeling and forecasting. It is argued that this gives models and predictions which better reflect reality. The SDE approach also offers a more adequate framework for modeling and a number of efficient tools for model building. A software package (CTSM-R) for SDE-based modeling is briefly described.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The integral over the multivariate Gaussian density is \(\frac{1}{\sqrt{\left( 2\pi \right) ^k}\vert \Sigma \vert } \int e ^ {\left( -\frac{1}{2} \left( \mathbf {x} - \mathbf {\mu }\right) ^T \Sigma ^{-1} \left( \mathbf {x} - \mathbf {\mu }\right) \right) } d\mathbf {x} = 1\).
References
Burnham, K., Anderson, D.: Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer (2002)
CTSM-R (Continuous Time Stochastic Modelling in R). www.ctsm.info
DIACON Project. www.diacongroup.org. New Technologies for treatment of Type 1 diabetes
Donnet, S., Samson, A.: A review on estimation of stochastic differential equations for pharmacokinetic/pharmacodynamic models. In: Advanced Drug Delivery Reviews (2013). doi:10.1016/j.addr.2013.03.005. http://www.sciencedirect.com/science/article/pii/S0169409X13000501
Duun-Henriksen, A., Juhl, R., Schmidt, S., Nørgaard, K., Madsen, H.: Modelling the effect of exercise on insulin pharmacokinetics in “continuous subcutaneous insulin infusion” treated type 1 diabetes patients. Technical report DTU Compute-Technical Report-2013, Technical University of Denmark (2013)
Jazwinski, A.H.: Stochastic processes and flitering theory. Dover publications, Inc. (1970)
Jellife, R., Schumitzky, A., Van Guilder, M.: Population pharmacokinetics/pharmacodynamics modeling; parametric and nonparametric methods. Ther. Drug Monit. 22, 354–365 (2000)
Karlsson, M., Beal, S., Sheiner, L.: Three new residual error models for population pk/pd analysis. J. Pharmacokinet. Pharmacodyn. 23, 651–672 (1995)
Klim, S., Mortensen, S.B., Kristensen, N.R., Overgaard, R.V., Madsen, H.: Population stochastic modelling (PSM)-an R package for mixed-effects models based on stochastic differential equations. Comput. Methods Progr. Biomed. 94(3), 279–289 (2009). doi:10.1016/j.cmpb.2009.02.001. http://www.sciencedirect.com/science/article/pii/S0169260709000455
Kristensen, N.R., Madsen, H.: Continuous time stochastic modelling–CTSM 2.3 Mathamatics guide. Technical University of Denmark, DTU Informatics, Building 321 (2003). www.ctsm.info
Kristensen, N.R., Madsen, H., Jørgensen, S.B.: A method for systematic improvement of stochastic gray-box models. Comput. Chem. Eng. 116, 1431–1449 (2004)
Kristensen, N.R., Madsen, H., Jørgensen, S.B.: Parameter estimation in stochastic grey-box models. Automatica 40(2), 225–237 (2004)
Lindsey, J., Jones, B., Jarvis, P.: Some statistical issues in modelling pharmacokinetic data. Stat. Med. 20, 2775–2783 (2001)
Löwe, R., Mikkelsen, P., Madsen, H.: Stochastic rainfall-runoff forecasting: parameter estimation, multi-step prediction, and evaluation of overflow risk. Stoch. Environ. Res. Risk Assess. 28(3), 505-516 (2014). doi:10.1007/s00477-013-0768-0. (Offprint, no public access)
Lv, D., Breton, M.D., Farhy, L.S.: Pharmacokinetics modeling of exogenous glucagon in Type 1 diabetes mellitus patients. Diabet. Technol. Ther. 15(11), 935–941 (2013). doi:10.1089/dia.2013.0150. http://online.liebertpub.com.globalproxy.cvt.dk/doi/abs/10.1089/dia.2013.0150
Madsen, H.: Time Series Analysis. Chapman and Hall (2008)
Madsen, H., Thyregod, P.: Introduction to general and generalized linear models. Chapman and Hall (2011)
Møller, J., Madsen, H.: Stochastic state space modelling of nonlinear systems–with application to marine ecosystems. In: IMM-PHD-2010-246. Technical University of Denmark, DTU Informatics, Building 321 (2010)
Møller, J., Phillipsen, K.R., Christensen, L.E., Madsen, H.: Development of a restricted state space stochastic differential equation model for bacterial growth in rich media. J. Theor. Biol. 305, 78–87 (2012). doi:10.1016/j.jtbi.2012.04.015
Nielsen, H.A., Madsen, H.: A generalization of some classical time series tools. Comput. Stat. Data Anal. 37, 13–31 (2001)
Pawitan, Y.: In all likelihood: Statistical modelling and inference using likelihood. Oxford Science Publications (2001)
Philipsen, K.R., Christiansen, L.E., Hasman, H., Madsen, H.: Modelling conjugation with stochastic differential equations. J. Theor. Biol. 263(1), 134–142 (2010). doi:10.1016/j.jtbi.2009.11.011
Schmidt, S., Finan, D.A., Duun-Henriksen, A.K., Jørgensen, J.B., Madsen, H., Bengtsson, H., Holst, J.J., Madsbad, S., Nørgaard, K.: Effects of everyday life events on glucose, insulin, and glucagon dynamics in continuous subcutaneous insulin infusion treated type 1 diabetes: Collection of clinical data for glucose modeling. Diabet. Technol. Ther. 4(3), 210–217 (2012). doi:10.1089/dia.2011.0101. http://online.liebertpub.com/doi/abs/10.1089/dia.2011.0101
Tornøe, C.W., Agersø, H., Jonsson, E.N., Madsen, H., Nielsen, H.A.: Non-linear mixed-effects pharmacokinetic/pharmacodynamic modelling in nlme using differential equations. Computer Methods and Programs in Biomedicine. Comput. Methods Progr. Biomed. 76(1), 31–40 (2004). doi:10.1016/j.cmpb.2004.01.001
Tornøe, C.W., Jacobsen, J., Pedersen, O., Hansen, T., Madsen, H.: Grey-box modelling of pharmacokinetic/pharmacodynamic systems. J. Pharmacokinet. Pharmacodyn. 31(5), 401–417 (2004)
Wilinska, M.E., Chassin, L.J., Acerini, C.L., Allen, J.M., Dunger, D.B., Hovorka, R.: Simulation environment to evaluate closed-loop insulin delivery systems in type 1 diabetes. J. Diabet. Sci. Technol. 4(1), 132–144 (2010)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix A: Extended Kalman Filtering
Appendix A: Extended Kalman Filtering
For nonlinear models the innovation vectors \({\mathbf {\varepsilon }_k}\) (or \({\mathbf {\varepsilon }^i_k}\)) and their covariance matrices \({{\Sigma }^{yy}_{k|k-1}}\) (or \({{\Sigma }^{yy,i}_{k|k-1}}\)) can be computed recursively by means of the extended Kalman filter (EKF) as outlined in the following.
Consider first the linear time-varying model
in the following we will use \(\mathbf {A}(t)\), \(\mathbf {B}(t)\), and \(\mathbf {\sigma }(t)\) as short-hand notation for \(\mathbf {A}(\mathbf {u}_t,t,\mathbf {\theta })\), \(B(\mathbf {u}_t,t,\mathbf {\theta })\), and \(\mathbf {\sigma }(\mathbf {u}_t,t,\mathbf {\theta })\).
We will restrict ourselves to the initial value problem; solve (46) for \(t\in [t_k,t_{k+1}]\) given that the initial condition \(X_{t_k}\sim N(\hat{\mathbf {x}}_{k|k}, \Sigma ^{xx}_{k|k})\). This is the kind of solution we would get from the ordinary Kalman filter in the update step.
Now if we consider, the transformation
then by Itô’s Lemma, it can be shown that the process \(\mathbf {Z}_t\) is governed by the Itô stochastic differential equation
with initial conditions \(\mathbf {Z}_{t_k}\sim N(\hat{\mathbf {x}}_{k|k}, \Sigma ^{xx}_{k|k})\). The solution to (49) is given by the integral equation
Now inserting the inverse of the transformation (48) gives
Taking the exception and variance on both sides of (51) gives
where we have used Itô isometry in the second equation for the variance. Now differentiation the above expression w.r.t. time gives
with initial conditions given by \(E[\mathbf {X}_{t_k}]=\hat{\mathbf {x}}_{k|k}\) and \(V[\mathbf {X}_{t_k}]=\Sigma ^{xx}_{k|k}\).
For the nonlinear case
we introduce the Jacobian of \(\mathbf {f}\) around the expectation of \(\mathbf {X}_t\) (\(\hat{\mathbf {x}}_t=E[\mathbf {X}_t]\)), we will use the following short hand notation
where \(\hat{\mathbf {x}}_{t}\) is the expectation of \(\mathbf {X}_t\) at time t, this implies that we can write the first-order Taylor expansion of (56) as
Using the results from the linear time-varying system above, we get the following approximate solution to the (59)
with initial conditions \(E[\mathbf {X}_{t_k}]=\hat{\mathbf {x}}_{k|k}\) and \(V[\mathbf {X}_{t_k}]=\Sigma ^{xx}_{k|k}\). Equations (60) and (61) constitute the basis of the prediction step in the Extended Kalman Filter, which for completeness is given below
Theorem 1
(Continuous-discrete time extended Kalman filter) With given initial conditions for the \({\hat{\mathbf {x}}_{1|0}=\mathbf {x}_0}\) and \({{\Sigma }^{xx}_{1|0}= {\Sigma }^{xx}_0}\) the extended Kalman filter approximations are given by; the output prediction equations:
the innovation and Kalman gain equation:
the updating equations:
and the state prediction equations:
where the following short-hand notation has been applied:
The prediction step was covered above and the updating step can be derived from linearization of the observation equation and the projection theorem [6]. From the construction above, it is clear that the approximation is only likely to hold if the nonlinearities are not too strong. This implies that the sampling frequency is fast enough for the prediction equations to be a good approximation and that the accuracy in the observation equation is good enough for the Gaussian approximation to hold approximately. Even though “simulation” through the prediction equations is available in CTSM-R, it is recommended that simulation results are verified (or indeed performed), by real-stochastic simulations (e.g., by simple Euler simulations).
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Juhl, R., Møller, J.K., Jørgensen, J.B., Madsen, H. (2016). Modeling and Prediction Using Stochastic Differential Equations. In: Kirchsteiger, H., Jørgensen, J., Renard, E., del Re, L. (eds) Prediction Methods for Blood Glucose Concentration. Lecture Notes in Bioengineering. Springer, Cham. https://doi.org/10.1007/978-3-319-25913-0_10
Download citation
DOI: https://doi.org/10.1007/978-3-319-25913-0_10
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-25911-6
Online ISBN: 978-3-319-25913-0
eBook Packages: EngineeringEngineering (R0)