Abstract
The paper presents a novel algorithm in 3D computed tomography (CT) dedicated to accurate region of interest (ROI) reconstruction. Here image data acquisition involves X-ray sources positioned on any piecewise smooth 3D-curve satisfying the very generic, Tuy’s conditions and uses only X-rays passing through the ROI. Our ROI reconstruction algorithm implements an iterative procedure where we systematically alternate intermediary reconstructions by Grangeat’s cone-beam inversion formula with density smoothing by shrinkage of wavelets coefficients. We validate the accuracy of our ROI reconstruction algorithm for a 3D Shepp-Logan phantom and a 3D image of a human jaw, and a curve composed of two (discretized) orthogonal circles by simulating ROI censored X-ray acquisition and our iterative ROI reconstruction for a family of spherical ROIs of various radii. The main result is that provided the ROI radius is larger than a critical radius, our procedure converges in less than 23 iterations in most cases, and that our ROI reconstruction remains reasonably accurate.
| Original language | English |
|---|---|
| Title of host publication | Fully3D Proceedings 2015 |
| Pages | 618-621 |
| Publication status | Published - 2015 |
| Externally published | Yes |
| Event | 13th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine - Duration: 1 Jun 2015 → … |
Conference
| Conference | 13th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine |
|---|---|
| Period | 1/06/15 → … |