Multiscale cone-beam x-ray reconstruction

Abstract

We address the problem of reconstructing a three-dimensional volume from a set of two-dimensional X-ray projections. We present a time efficient solution based on a multiscale estimation technique. Estimation is first performed at a coarse resolution. Then the resolution is increased step by step and at each step a new estimation is performed, using an initial value derived from the volume estimated at the preceding level of resolution. The method is illustrated by results obtained on geometric and anatomic phantoms.

3D medical image coding using separable 3D wavelet decomposition and lattice vector quantization

“…The 2D projections are processed towards the reconstruction algorithms [32,37] and the reconstructed 3D image is displayed with a dedicated software. The system can generate up to 512 3 discrete volumes with isotropic voxels of 356 microns, each voxel coded on 12 bits.…”

Nonseparable wavelet-based cone-beam reconstruction in 3-d rotational angiography

“…However, due to the limited number of available projections in RA, the raw reconstructed slices are quite noisy. Moreover, we may notice that the information content in the reconstructed volume is restricted to a very limited support, around 5% in the case of vascular imaging [10]. It is then natural to ask for a multiscale representation of the 3-D object in order to achieve wavelet processing such as noise reduction, local wavelet synthesis.…”