Parameter estimation for a model of blood volume change during hemodialysis

Parameter estimation was performed for a model of blood volume change during hemodialysis. A sensitivity analysis of the effects of model parameter variation on the predicted blood volume change showed that three of eleven model parameters should be estimated. A Bayesian parameter estimator was used with data collected from fourteen maintenance hemodialysis patients. Initial results indicated that events at the beginning of hemodialysis were not accounted for in the model. Inclusion of the effects of a proposed capillary pressure fall in the model improved the model fit.<<ETX>>