Modeling the residue function in DSC‐MRI simulations: Analytical approximation to in vivo data

Abstract

A bi-exponential model should therefore be used in future numerical simulations of DSC-MRI instead of the exponential function.

Perfusion deconvolution in DSC-MRI with dispersion-compliant bases

“…The residue function then decreases during time with the clearance of the tracer. A recent work (Mehndiratta et al, 2014b) shows that the shape of r(t) is best described in vivo by a bi-exponential model, a finding in agreement with previous literature (Park and Payne, 2013). This model accounts for fast and slow flowing capillary components…”

“…(6) with each of the various dispersion kernels: gamma (GDK), exponential (EDK), and lognormal (LNDK). For the bi-exponential r(t), we chose the median normal tissue parameters given by Mehndiratta et al (2014b): f = 0.97, τ F = 0.68 and τ S = 0.05. For the dispersion kernel parameters we adopted the values corresponding to low, medium, and high dispersion (table 1).…”

Robustness of spatio‐temporal regularization in perfusion MRI deconvolution: An application to acute ischemic stroke

“…Each voxel in the 2D (or 3D) spatial domain is associated to a one‐dimensional signal representing the impulse response function of this voxel. Different models representing potential shapes for the tissue impulse response function have been proposed in the literature . In our simulator, the choice between a box‐shaped, triangular or single exponential first‐order model is given: …”

“…Each voxel v in the 2D (or 3D) spatial domain is associated to a one-dimensional signal f v ðtÞ representing the impulse response function of this voxel. Different models representing potential shapes for the tissue impulse response function have been proposed in the literature (8,9,22). In our simulator, the choice between a boxshaped, triangular or single exponential first-order model is given: ½f…”

Correcting for large vessel contamination in dynamic susceptibility contrast perfusion MRI by extension to a physiological model of the vasculature

“…In principle, the contribution of the MV component to the signal should match that of the AIF, ignoring dispersive effects and complex partial volume effects ; thus, it should be possible to use the existing information about the measured AIF to assist in MV correction. In this work, we propose a model‐based solution to the correction for MV contamination inspired by related recent work in Arterial Spin Labeling , building upon work previously presented in abstract form . To do so, we build upon the vascular model of Mouridsen et al for the DSC‐MRI tissue signal, by incorporating a separate MV component, which must be estimated simultaneously from the DSC‐MRI data.…”