Rapidly-converging multigrid reconstruction of cone-beam tomographic data

TL;DR

This paper introduces a fast, robust iterative reconstruction method for large-angle cone-beam CT that uses a space-filling trajectory and multigrid techniques to achieve rapid convergence on large datasets.

Contribution

The paper proposes a novel multigrid reconstruction scheme utilizing a space-filling trajectory and an approximate deconvolution pre-conditioner for efficient large-scale CBCT.

Findings

Achieves rapid convergence compared to traditional methods

Demonstrates robustness with large-angle, large-dataset CBCT data

Validates effectiveness through comparative experiments

Abstract

In the context of large-angle cone-beam tomography (CBCT), we present a practical iterative reconstruction (IR) scheme designed for rapid convergence as required for large datasets. The robustness of the reconstruction is provided by the "space-filling" source trajectory along which the experimental data is collected. The speed of convergence is achieved by leveraging the highly isotropic nature of this trajectory to design an approximate deconvolution filter that serves as a pre-conditioner in a multi-grid scheme. We demonstrate this IR scheme for CBCT and compare convergence to that of more traditional techniques.