Related Experiment Video
Updated: Mar 11, 2026
08:41
Patient-Specific Polyvinyl Alcohol Phantom Fabrication with Ultrasound and X-Ray Contrast for Brain Tumor Surgery Planning
Published on: July 14, 2020
9.2K
Comparison of low-contrast detectability between two CT reconstruction algorithms using voxel-based 3D printed
Justin Solomon1, Alexandre Ba2, François Bochud2
1Department of Radiology, Carl E. Ravin Advanced Imaging Laboratories, Clinical Imaging Physics Group, Duke University Medical Center, Durham, North Carolina 27705.
Medical Physics
|December 3, 2016
Summary
Novel 3D printed phantoms show that SAFIRE (sinogram affirmed iterative reconstruction) improves low-contrast detectability over filtered-backprojection (FBP). However, SAFIRE’s dose reduction potential varies with background texture complexity.
Area of Science:
- Medical Imaging Physics
- Image Reconstruction Algorithms
- Radiological Phantom Development
Background:
- Accurate low-contrast detectability is crucial in CT imaging.
- Anatomical noise (background texture) can impact image quality and algorithm performance.
- Iterative reconstruction algorithms offer potential dose reduction but require validation.
Purpose of the Study:
- To compare low-contrast detectability between FBP and SAFIRE using 3D printed textured phantoms.
- To evaluate the impact of background texture complexity on SAFIRE's dose reduction potential.
Main Methods:
- Voxel-based 3D printed phantoms with realistic liver-like textures were created.
- Low-contrast detectability was assessed using a channelized Hotelling observer across varying doses and contrasts.
- Multivariate regression analyzed the effects of dose, reconstruction algorithm, and background type on detectability.
Main Results:
- Detectability increased with dose and contrast, and was higher in uniform backgrounds.
- SAFIRE demonstrated higher detectability (d') compared to FBP across all conditions.
- SAFIRE's estimated dose reduction potential ranged from 8% to 27%, depending on background texture complexity.
Conclusions:
- SAFIRE consistently improves detectability over FBP, irrespective of background texture.
- The effectiveness of SAFIRE in reducing dose is significantly influenced by background texture complexity.
- Realistic phantom models are essential for evaluating iterative reconstruction algorithms.
