Testing a model for MR imager noise

This work investigates the imager-induced noise components in complex-valued MRI signals. We present background for a noise model, and address the task of validating this model. We identify certain aspects of MRI data which potentially bias the results of goodness-of-fit tests, and suggest a testing formulation that accounts for these issues. This formulation is then applied to sets of echo-planar image sequences, successfully verifying consistency with the proposed model.