Evaluation of model-independent deconvolution techniques to estimate blood perfusion

This report evaluates several methods to estimate blood perfusion and residue functions in dynamic contrast enhanced (DCE) MRI. Among these are model-dependent and model-independent techniques. All methods were applied to series of Monte Carlo simulations to evaluate the accuracy in order to reproduce different underlying vascular residue functions and blood perfusions. Of the model-independent approaches the use of B-splines with Tikhonov regularization was shown to have a reasonable accuracy in blood perfusion estimations and was less biased than all model-dependent approaches. This technique seems most promising for application to experimental data.