Related Experiment Video
Updated: Jul 31, 2026
05:05
Four-Dimensional CT Analysis Using Sequential 3D-3D Registration
Published on: November 23, 2019
A differentiable Shepp-Logan phantom and its applications in exact cone-beam CT
Hengyong Yu1, Shiying Zhao, Ge Wang
1CT/Micro-CT Laboratory, Department of Radiology, University of Iowa, Iowa City, 52242, USA. hengyong-yu@uiowa.edu
Physics in Medicine and Biology
|November 25, 2005
Summary
Researchers developed a differentiable Shepp-Logan phantom (DSLP) to improve cone-beam CT algorithm evaluation. The DSLP offers better image quality and complements the traditional phantom for validating reconstruction methods.
Area of Science:
- Medical Imaging
- Computational Science
Background:
- Exact cone-beam CT reconstruction algorithms address the long object problem.
- The traditional 3D Shepp-Logan phantom (SLP) has discontinuities limiting its use for validating these algorithms.
Purpose of the Study:
- To develop a differentiable Shepp-Logan phantom (DSLP) for more accurate numerical simulations.
- To evaluate the performance of exact cone-beam CT algorithms using the new DSLP.
Main Methods:
- Constructed a differentiable polynomial function to approximate the unit rectangular function.
- Developed a differentiable ellipsoid phantom with a differentiable x-ray transform.
- Proposed the differentiable Shepp-Logan phantom (DSLP) for algorithm simulation.
Main Results:
- The DSLP is differentiable for any smooth scanning trajectory.
- Numerical simulations demonstrated superior image quality with the reconstructed DSLP compared to the SLP.
- The DSLP proved complementary to the traditional SLP for algorithm evaluation.
Conclusions:
- The proposed DSLP enhances the numerical simulation of exact cone-beam CT algorithms.
- DSLP provides a more robust phantom for evaluating algorithm performance and image quality.
