| Summary: | The systematic investigation of susceptibility-induced contrast in MRI is important to improve our understanding of the influence of tissue microstructure on dynamic susceptibility contrast (DSC)-MRI derived perfusion data. The Finite Perturber Method (FPM) has previously been used to investigate susceptibility contrast in MRI arising from arbitrarily shaped structures. However, the FPM has low computational efficiency in simulating water diffusion, especially for complex three-dimensional structures that mimic tissue. In this work, an improved computational approach that combines the FPM with a matrix-based finite difference method (FDM), termed the Finite Perturber Finite Difference Method (FPFDM), was developed to more efficiently investigate the biophysical basis of DSC-MRI data and its sensitivity to vascular geometry and contrast agent (CA) distribution within tissue. The application of the FPFDM to the physiological and physical conditions encountered in a typical DSC-MRI brain tumor study enabled the derivation of a new DSC-MRI metric, termed the Transverse Relaxivity at Tracer Equilibrium (TRATE), which we propose specifically reports on tumor cellular properties. Computational FPFDM studies revealed that TRATE depends on cellular density, size, shape and spatial distribution. To validate the in vivo sensitivity of TRATE it was evaluated in two animal brain tumor models, the 9L and C6, which have varying cellular characteristics. The TRATE values were also compared to measures of the apparent diffusion coefficient (ADC), the CA transfer constant (Ktrans), the extravascular extracellular volume fraction (ve) and histological data. The TRATE values in 9L tumors were significantly higher than those in C6 tumors, a finding that reflects the histologically confirmed higher cell density in 9L tumors and lower cellular density. A voxel-wise comparison of TRATE with ADC, ve, and Ktrans maps showed low spatial correlation, indicating it is providing unique and complementary information on tumor status. In summary, the studies described herein highlight the value of pairing computational and experimental advancements in order to enhance our characterization of DSC-MRI contrast mechanisms and how such mechanisms can be leveraged to derive new non-invasive metrics for evaluating brain tumors.
|
|---|
