The effective parameters on precision analysis of Dynamic Magnetic Resonance Imaging data

Introduction: Precision analysis of the estimated kinetic parameters from Dynamic ContrastEnhanced –Magnetic Resonance images (DCE-MRI) is very important when they are used asa measure for diagnosis and therapy evaluation. In this work, the effect of the coincidencebias in two major effective parameters including pre contrast longitudinal relaxation time(T10), flip angle (α) on the error of kinetic parameter estimation in brain DCE-MRI studieswas evaluatedMethods and Materials: In this study the effect of the simultaneous bias in T10 and α wasanalytically evaluated. Thus, the extended Toft’s model was used to simulate 3150 contrastconcentration curves. Using the previously published equation, the bias in T10 and α wereadded to simulated curves and then kinetic parameters of Toft’s model were estimated for allbiased and non-biased simulated profiles. Finally, mean percentage error (MPE) as criteriafor evaluation of accuracy of estimated parameters was calculatedResults: The results of this study indicated that simultaneous incorporation of flip angle andT10 bias may induce a large uncertainty in CA concentration profiles and consequentlycauses a significant bias in estimated kinetic parameters. For example, for 20% relative biasin each α and T10, mean percentage error of computed parameters are as following: MPE ofvp is around 30%, MPE of Ktrans is almost 40% and MPE of kep is 80%Discussion and Conclusion: In conclusion, we found that α and T10 deviations couldconsequence important errors in plasma volume estimation of brain tissues through Toftsmodel fitting, while estimation of transfer constants could be more reliable and robust inpresence of deviation in flip angle or pre-contrast longitudinal relaxation time in brain DCEMRIstudies.

کلیدواژه ها:

Pharmacokinetic Analysis ، Contrast Concentration Bias ، Accuracy of kinetic parameters