Movement correction of the kidney in dynamic MRI scans using FFT phase difference movement detection

Abstract

To measure cortical and medullary MR renograms, regions of interest (ROIs) are placed on the kidney in images acquired using dynamic MRI. Since native kidneys move with breathing, and breath-holding techniques are not feasible, movement correction is necessary. In this contribution we compare three correction methods, based on image matching, phase difference movement detection (PDMD), and cross-correlation, respectively. The PDMDbased method showed the best performance and was able to determine kidney movement in our test series in 68% of the scans with no visible deviation, and in 88% of the scans if a one-pixel deviation is considered acceptable.

Multislice perfusion of the kidneys using parallel imaging: Image acquisition and analysis strategies

“…Recently, realignment of renal dynamic susceptibility contrast cerebral blood flow (CBF) data (22,23) has been suggested. Here we used a six-parameter rigid-body FMRIB's linear image registration tool (FLIRT) registration algorithm to correct kidney movement.…”

“…Free respiration predominantly causes a positional shift and little deformation of the shape of the kidney (22), and as such rigid body realignment algorithms may correct respiratoryinduced motion, particularly when multislice ASL data are available. Realignment methods are currently being assessed to correct dynamic contrast-enhanced MRI kidney time series (22,23) but have not been applied to ASL time series. Realignment provides the advantages of retaining all images in the data analysis, as opposed to data sorting, and minimizes movement-induced reduction in perfusion signal likely to remain for BS methods.…”

Improvement of MRI‐functional measurement with automatic movement correction in native and transplanted kidneys

“…[1] is a similarity criterion, computed on gray levels intensities, between the part of a reference image containing the kidney and the area it overlays in the image to register (14,16). A number of similarity criteria have been proposed in the literature (4).…”

“…For example, registration problems involving pure translation can be recovered by computing a phase difference in the Fourier domain (14). Methods have been proposed to allow subpixel accuracy (17), and estimation of more complex displacements (a log-polar transformation is applied to the magnitude spectrum and the rotation and scale is recovered by computing phase difference in the logpolar space (18,19)).…”

“…Methods have been proposed to allow subpixel accuracy (17), and estimation of more complex displacements (a log-polar transformation is applied to the magnitude spectrum and the rotation and scale is recovered by computing phase difference in the logpolar space (18,19)). Fourier-based approaches are very low time-consuming and also more robust to estimate translation motion than methods based on gray levels intensity (see (14)). …”