Consistent projection reconstruction (CPR) techniques for MRI

714

733

The reconstruction of artifact-free images from radially encoded MRI acquisitions poses a difficult task for undersampled data sets, that is for a much lower number of spokes in k-space than data samples per spoke. Here, we developed an iterative reconstruction method for undersampled radial MRI which (i) is based on a nonlinear optimization, (ii) allows for the incorporation of prior knowledge with use of penalty functions, and (iii) deals with data from multiple coils. The procedure arises as a twostep mechanism which first estimates the coil profiles and then renders a final image that complies with the actual observations. Prior knowledge is introduced by penalizing edges in coil profiles and by a total variation constraint for the final image

Section: Regularization By a Priori Informationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Undersampled radial MRI with multiple coils. Iterative image reconstruction using a total variation constraint

Block

Uecker

Frahm

2007

714

733

“…Next, the results from the weighted least-squares inversion are used in Eq. [11] to yield the rotational motion over time. This rotational motion time record is median-filtered and low-pass-filtered to reduce noise.…”

Section: Correction Algorithmmentioning

confidence: 99%

“…Some techniques simply try to minimize the effect of motion with approaches such as respiratory ordered view angles (1) (analogous to respiratory ordered phase encoding (9) used in conjunction with 2DFT imaging), or with fast scanning (10). Some techniques impose consistency constraints on the PR data to filter out (11) or resort (12) inconsistent data. In (12), Gai and Axel use Ludwig-Helgason consistency conditions to correct linogram and projection reconstruction data for in-plane scaling, rotation, and translation of the imaged object.…”

mentioning

confidence: 99%

Self‐navigated motion correction using moments of spatial projections in radial MRI

Welch¹,

Rossman²,

Felmlee³

et al. 2004

Interest in radial MRI (also known as projection reconstruction (PR) MRI) has increased recently for uses such as fast scanning and undersampled acquisitions. Additionally, PR acquisitions offer intrinsic advantages over standard two-dimensional Fourier transform (2DFT) imaging with respect to motion of the imaged object. It is well known that aligning each spatial domain projection's center of mass (calculated using the 0th and 1st moments) to the center of the field of view (FOV) corrects shifts caused by in-plane translation. In this work, a previously unrealized ability to determine the in-plane rotational motion of an imaged object using the 2nd moments of the spatial domain projections in conjunction with a specific projection angle acquisition time order is reported. We performed the correction using only the PR data itself acquired with the newly proposed projection angle acquisition time order. With the proposed view angle acquisition order, the acquisition is "self-navigating" with respect to both in-plane translation and rotation. We reconstructed the images using the aligned projections and detected acquisition angles to significantly reduce image artifacts due to such motion. MR acquisitions using a radial MR trajectory and projection reconstruction (PR) are known to have intrinsic advantages over two-dimensional Fourier transform (2DFT) k-space trajectories for imaging a moving object (1). Motion artifacts in PR manifest as radial streaks perpendicular to the direction of motion, with diminished amplitude near the moving object. In 2DFT approaches, motion often creates ghosts in one direction, with the strongest intensity near the source of movement. Because of the improved motion characteristics of PR methods, as well as other properties, interest in PR has increased recently for applications such as high-speed 3D imaging (2), MR angiography (MRA) (3,4), dynamic imaging and fluoroscopy (5,6), catheter tracking (7), and reduced field of view (FOV) imaging (8).Many techniques have been proposed to further improve the robustness of PR against motion artifacts. Some techniques simply try to minimize the effect of motion with approaches such as respiratory ordered view angles (1) (analogous to respiratory ordered phase encoding (9) used in conjunction with 2DFT imaging), or with fast scanning (10). Some techniques impose consistency constraints on the PR data to filter out (11) or resort (12) inconsistent data. In (12), Gai and Axel use Ludwig-Helgason consistency conditions to correct linogram and projection reconstruction data for in-plane scaling, rotation, and translation of the imaged object. Using the projections' observed 2nd moment values, the approach by Gai and Axel attempts to resort projections to partially compensate for limited rotations occurring in the monotonic segments of the 2nd moment trajectory versus projection view angle. Their approach and that of others (13) compensates for the effect on in-plane translation by shifting projections by their centers of mass. Another method corre...

“…This approach, however, requires significant additional hardware and accurate calibration between two modalities. Alternately, the ''self-navigating'' properties of radial (11)(12)(13), spiral (14) and periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) (15) acquisition schemes have been proposed for motion correction. In these methods, the central part of k-space is usually oversampled, allowing for motion detection.…”

Section: Introductionmentioning

confidence: 99%

High temporal resolution retrospective motion correction with radial parallel imaging

Wei

Huang

Duensing

2011

A method for motion correction in multicoil imaging applications, involving both data collection and reconstruction, is presented. A bit-reversed radial acquisition scheme, in conjunction with a rapid self-calibrated parallel imaging method, Generalized auto-calibrating partial parallel acquisition (GRAPPA) operator for wider radial bands (GROWL), is used to achieve motion correction at a high temporal resolution. View-by-view in-plane motion correction is achieved in 2D imaging, while 3D motion correction is achieved for every two consecutive slice-encoding planes in 3D imaging. In the proposed technique, GROWL contributes in two aspects: First, a central k-space circle/cylinder used as the motion-free reference is generated from a small number of radial lines/planes; Second, undersampled k-space regions resulting from rotation and inconsistent (e.g. intraview and nonrigid body) motion can be filled in. When compared with navigator-based motion correction methods, the proposed method does not prolong scan time and can be applied to short-TR sequences.