Cerebral perfusion maps from dynamic contrast MRI data utilizing Rician statistics

Bolus tracking of contrast agent with MRI is a well established technique for measurement of local cerebral hemodynamic parameters flow, volume and mean transit time. When performed on a voxel-by-voxel basis, it allows development of hemodynamic parameter maps useful for assessment of ischemic damage following stroke and tumor characterization in cancer. The analysis of the acquired dynamic data requires the use of deconvolution to reconstruct the residue function (R) of the contrast agent. Measurement of the tissue time course and the arterial input function are obtained by T2 or T2* weighted sequences. Reconstruction of R provides estimates of flow, volume and mean transit time. The raw MRI scan signal intensity is well approximated by Rician statistics. The standard approach to estimation involves logarithmic transformation and least squares deconvolution. At low signal to noise ratio this approach is not efficient and as an alternative this work adopts an iterative re-weighted non-linear least squares (IRWNLLS) algorithm to incorporate Rician statistics, impose constraints on the residue function and optimize for tracer arrival delay. The algorithm is implemented on a voxel-by-voxel basis and cerebral maps for the hemodynamic parameters flow, volume and mean transit time are presented. In addition, an automatic segmentation technique which takes into account both spatial and temporal variation is presented. This segmentation technique is shape driven, choosing only voxels that correlate highly with a well-known arterial input function template.