Dynamic contrast enhanced (DCE) and dynamic susceptibility contrast (DSC) MRI Dynamic contrast enhanced (DCE) and dynamic susceptibility contrast (DSC) MRI, which can be acquired within routine clinical MRI sessions with only a few extra minutes, have been shown to be of great clinical value for diagnosis, staging, therapy guidance and assessment of treatment response for many diseases. However, the DCE and DSC images need to be post-processed by pharmacokinetic modeling to extract parameters for clinical interpretation. Given that hemodynamics varies with organs and/or diseases, no one pharmacokinetic model can meet all needs. Most commercial software solutions rarely provide more than one model, and thus could compromise the results if the available model does not meet the clinical or imaging acquisition condition. A research group in the Radiology and Medical School of University of Michigan has reported a software implementation that provides six different pharmacokinetic models to meet broader needs for quantification of DCE and DSC images.
Integral Implementation of Multiple Pharmacokinetic Models The proposed software implementation (written in C++ and QT) consists of six pharmacokinetic models within a common framework for quantifying blood flow, blood volume, vascular permeability, and extravascular extracellular space, which are applicable to multiple body sites including, brain, neck, liver, lung, kidney, cervix, prostate, rectum and pelvis. Multiple numerical computation methods are used to deal with computation speed, stability of extracted parameters, low signal-noise ratio (or contrast-noise ratio) of DCE (or DSC) images, and limited temporal resolution. The implementation also allows a user to quantify DCE/DSC images within a volume of interest, either voxel by voxel or region by region.
Applications • DCE and DSC MRI image quantification. • Potential use in diagnosis, therapy guidance and treatment assessment.
Advantages • Multiple pharmacokinetic models are available to meet the organ/disease condition. • Optimized for computation speed and stability of extracted parameters.