Meeting Banner
Abstract #2339

Fully Bayesian Multi-model Inference for Parameter Estimation in DCE-MRI

Tammo Rukat 1 and Stefan A Reinsberg 1

1 Department of Physics and Astronomy, University of British Columbia, Vancouver, British Columbia, Canada

A fully Bayesian model mixing method for the estimation of haemodynamic parameters from DCE-MRI is being assessed. In particularly we examine the capability of weighing models of different complexity, such that the resulting parameter can be expected to be more accurate than the estimate from any single model. The Watanabe-Akaike information criterion (WAIC) is derived from the posterior likelihood distributions of the model parameters, which was sampled by adaptive MCMC. WAIC serves to calculate model mixing weights. This method is shown to be superior to the choice of any single model.

This abstract and the presentation materials are available to members only; a login is required.

Join Here