Correction of spatial distortion in magnetic resonance angiography for radiosurgical treatment planning of cerebral arteriovenous malformations

Schad, Lothar R.; Ehricke, Hans-Heino; Wowra, Berndt; Layer, G.; Engenhart, R.; Kauczor, H.-U.; Zabel, H.-J.; Brix, Gunnar; Lorenz, Walter J.

doi:10.1016/0730-725x(92)90012-o

Cited by 97 publications

(41 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Two early papers by Schad et al (15,16) discussed the "pincushion" effects seen with 2D phantoms. Subsequent schemes include using a stereotaxic frame as a reference marker (11,14), comparing phantom images from CT and MRI (7,19), and imaging specific MRI phantoms (1,13,17,18) that typically consist of an array of tubes filled with a suitable contrast agent.…”

mentioning

confidence: 99%

Use of spherical harmonic deconvolution methods to compensate for nonlinear gradient effects on MRI images

Janke

Zhao

Cowin

et al. 2004

Magnetic Resonance in Med

140

191

View full text Add to dashboard Cite

Spatial encoding in MR techniques is achieved by sampling the signal as a function of time in the presence of a magnetic field gradient. The gradients are assumed to generate a linear magnetic field gradient, and typical image reconstruction relies upon this approximation. However, high-speed gradients in the current generation of MRI scanners often sacrifice linearity for improvements in speed. Such nonlinearity results in distorted images. The problem is presented in terms of first principles, and a correction method based on a gradient field spherical harmonic expansion is proposed. In our case, the amount of distortion measured within a typical field of view (FOV) required for head imaging is sufficiently large that without the use of some distortion correction technique, the images would be of The quality of an MR image is dependent upon the accuracy with which the physical position is spatially encoded. Since MRI data are now used routinely for stereotaxy, longitudinal studies of atrophy, and functional studies, it is critical to ensure that images have no distortion or inhomogeneity. The principal machine-dependent sources of this inhomogeneity are eddy currents, gradient nonlinearity, and B 0 and B 1 inhomogeneity (14). In this study we present an analytical approach to calculating and removing the effects of nonlinear gradients only. We chose to focus on a gradient-only solution because of the recent interest in short-bore, high-speed gradients. Although peripheral nerve stimulation is a limiting feature of short rise times, such gradients have been found to be useful in high-speed echo-planar imaging (EPI) of the heart and diffusion tensor imaging (DTI) of the brain. To achieve short rise times and avoid peripheral nerve stimulation, designers have restricted the length of gradients and limited the number of turns in those gradients. Although these constraints are suitable for the implementation of pulse sequences with the desired speed, they have the undesired consequence of increased nonlinearity.Nonlinear pulsed field gradients induce image distortions due to incorrect spatial encoding of the signal. If we assume that the field gradients are linear, it follows that k-space is sampled linearly, and thus the fast Fourier transform (FFT) is suitable for reconstruction. However, any deviation from linearity in the gradients results in nonlinear data sampling and subsequent errors in image spatial encoding. A nonlinear FT would allow these data to be correctly transformed to an image. Unfortunately, nonlinear FT greatly increases computation time by N/log 2 N, relative to an FFT, which makes real-time image generation computationally prohibitive. Here we present a general analytical solution to correct image distortions induced by gradient nonlinearity. The method is applicable to any gradient configuration. It is robust and, more importantly, is based upon an approach whereby the FFT is maintained for image reconstruction.The notion of mapping and correcting such MRI distortions is not new. Two early pa...

show abstract

mentioning

confidence: 99%

Use of spherical harmonic deconvolution methods to compensate for nonlinear gradient effects on MRI images

Janke

Zhao

Cowin

et al. 2004

Magnetic Resonance in Med

140

191

View full text Add to dashboard Cite

show abstract

“…, 7 In this work, the spatial accuracy of the MR images was determined for the specific image sequences to be used to assess lung tumor and its motion. The fGRE images of the thoracic phantom demonstrated an acceptable level of spatial accuracy, because the displacements measured did not exceed 2.0 mm.…”

Section: Discussionmentioning

confidence: 99%

Assessment of geometrical accuracy of magnetic resonance images for radiation therapy of lung cancers

Koch

Liu

Olsson

et al. 2003

J Applied Clin Med Phys

View full text Add to dashboard Cite

The purpose of this research was to investigate the geometrical accuracy of magnetic resonance (MR) images used in the radiation therapy treatment planning for lung cancer. In this study, the capability of MR imaging to acquire dynamic two‐dimensional images was explored to access the motion of lung tumors. Due to a number of factors, including the use of a large field‐of‐view for the thorax, MR images are particularly subject to geometrical distortions caused by the inhomogeneity and gradient nonlinearity of the magnetic field. To quantify such distortions, we constructed a phantom, which approximated the dimensions of the upper thorax and included two air cavities. Evenly spaced vials containing contrast agent could be held in three directions with their cross‐sections in the coronal, sagittal, and axial planes, respectively, within the air cavities. MR images of the phantom were acquired using fast spin echo (FSE) and fast gradient echo (fGRE) sequences. The positions of the vials according to their centers of mass were measured from the MR images and registered to the corresponding computed tomography images for comparison. Results showed the fGRE sequence exhibited no errors >2.0 mm in the sagittal and coronal planes, whereas the FSE sequence produced images with errors between 2.0 and 4.0 mm along the phantom's perimeter in the axial plane. On the basis of these results, the fGRE sequence was considered to be clinically acceptable in acquiring images in all sagittal and coronal planes tested. However, the spatial accuracy in periphery of the axial FSE images exceeded the acceptable criteria for the acquisition parameters used in this study.PACS number(s): 87.57.–s, 87.61.–c

show abstract

“…The various distortion correction methods include SHDG, phase mapping, 20 adoption of a specially constructed phantom, 21 and the use of 2 frequency-encoding gradients. 22 We employed SHDG correction for nonlinearities of the gradientˆelds using the same method described by Jovicich and colleagues, 8 which employs information based on the design of the gradient coils; we used MRspeciˆc information, such as the gradient correction coe‹cients.…”

Section: Image Preprocessing For Vbmmentioning

confidence: 99%

Effects of Image Distortion Correction on Voxel-based Morphometry

Goto

Abe

Kabasawa³

et al. 2012

magn reson med sci

View full text Add to dashboard Cite

Purpose: We aimed to show that correcting image distortion signiˆcantly aŠects brain volumetry using voxel-based morphometry (VBM) and to assess whether the processing of distortion correction reduces system dependency.Materials and Methods: We obtained contiguous sagittal T 1 -weighted images of the brain from 22 healthy participants using 1.5-and 3-tesla magnetic resonance (MR) scanners, preprocessed images using Statistical Parametric Mapping 5, and tested the relation between distortion correction and brain volume using VBM.Results: Local brain volume signiˆcantly increased or decreased on corrected images compared with uncorrected images. In addition, the method used to correct image distortion for gradient nonlinearity produced fewer volumetric errors from MR system variation.Conclusion: This is theˆrst VBM study to show more precise volumetry using VBM with corrected images. These results indicate that multi-scanner or multi-site imaging trials require correction for distortion induced by gradient nonlinearity.

show abstract

Correction of spatial distortion in magnetic resonance angiography for radiosurgical treatment planning of cerebral arteriovenous malformations

Cited by 97 publications

References 14 publications

Use of spherical harmonic deconvolution methods to compensate for nonlinear gradient effects on MRI images

Use of spherical harmonic deconvolution methods to compensate for nonlinear gradient effects on MRI images

Assessment of geometrical accuracy of magnetic resonance images for radiation therapy of lung cancers

Effects of Image Distortion Correction on Voxel-based Morphometry

Contact Info

Product

Resources

About