Abstract:Non-uniformly sampled data in MRI applications must be interpolated onto a regular Cartesian grid to perform fast image reconstruction using FFT. The conventional method for this is gridding, which requires a density compensation function (DCF). The calculation of DCF may be time-consuming, ambiguously defined, and may not be always reusable due to changes in k-space trajectories. A recently proposed reconstruction method that eliminates the requirement of DCF is block uniform resampling (BURS) which uses singular value decomposition (SVD). However, the SVD is still computationally intensive. In this work, we present a modified BURS algorithm using conjugate gradient method (CGM) in place of direct SVD calculation. Calculation of a block of grid point values in each iteration further reduces the computational load. The new method reduces the calculation complexity while maintaining a high-quality reconstruction result. For an n-by-n matrix, the time complexity per iteration is reduced from O(n*n*n) in SVD to O(n*n) in CGM. The time can be further reduced when we stop the iteration in CGM earlier according to the norm of the residual vector. Using this method, the quality of the reconstructed image improves compared to regularized BURS. The reduced time complexity and improved reconstruction result make the new algorithm promising in dealing with large-sized images and 3D images.

A -Convergence Result and An Off-the-Grid Charge Algorithm for Curve Reconstruction in Inverse Problems

On the Convergence of the Accelerated Riccati Iteration Method.

Surface Reconstruction From Contours Based On Fast Calculation Of Distance Field

New Iterative Gridding Algorithm Using Conjugate Gradient Method

The Convergence of the Geometric Interpolation Algorithm.

Integrated Linear Reconstruction for Finite Volume Scheme on Arbitrary Unstructured Grids

An Integrated Linear Reconstruction for Finite Volume Scheme on Unstructured Grids

Nonlinear Fourier analysis for discontinuous conductivities: computational results

Convergence Analysis of the Algorithm in "Efficient and Robust Discrete Conformal Equivalence with Boundary"

A Note on the Convergence of Multigrid Methods for the Riesz–Space Equation and an Application to Image Deblurring

Fast convergence to non-isolated minima: four equivalent conditions for $\mathrm{C}^2$ functions

Convergence Analysis in Near-Field Imaging

A note on the convergence of multigrid methods for the Riesz-space equation and an application to image deblurring

Convergence analysis of inexact two-grid methods: A theoretical framework

Euclidean and Affine Curve Reconstruction

Curvature-Driven Conformal Deformations

Convergence of iterative algorithms for image reconstruction

Discrete Conformal Deformation: Algorithm and Experiments

Convergence studies on iterative algorithms for image reconstruction.

Algorithms and Convergence Results of Projection Methods for Inconsistent Feasibility Problems: A Review

Reconstruction of Random Fields Concentrated on an Unknown Curve using Irregularly Sampled Data