Dec 7 – 10, 2026
IPEN
America/Sao_Paulo timezone

Evaluation of Reconstruction Algorithms for Optical CT of Fricke Xylenol Gel Dosimeters

Dec 9, 2026, 11:00 AM
1h 40m
IPEN

IPEN

Cidade Universitária - Bairro Butantã CEP: 05508-000 - São Paulo - SP
Poster Nuclear Medicine Poster

Speaker

Victória Raposo

Description

Optical Computed Tomography (Optical CT) is used to obtain three-dimensional readouts of Fricke Xylenol Gel (FXG) dosimeters, where it is employed to map dose distributions in radiotherapy and nuclear medicine. In FXG dosimeters, ionizing radiation contributes to the oxidation of ferrous ions (Fe²⁺) into ferric ions (Fe³⁺), whose concentration is proportional to the absorbed dose. The concentration of ferric ions is proportional to the absorbed dose, and for measurements using an optical CT scanner, the dose can be related to the optical attenuation coefficients of FXG. However, the ferric ions diffuse through the gel matrix after irradiation, causing blurring of the recorded dose distribution and reducing spatial accuracy over time. This diffusion represents one of the main limitations of FXG dosimetry, making 3D reconstruction a critical step for the assessment of dose distribution.

The quality of the reconstructed images depends on the reconstruction algorithm used, which can affect how physical phenomena such as ferric ion diffusion are visualized, as well as spatial resolution, image noise, and reconstruction artifacts.
This study aims to compare different image reconstruction algorithms applied to Optical CT of FXG dosimeters, investigating how these methods affect the visualization and measurement of diffusion effects while maintaining the accuracy of the reconstructed dose distribution.
Particular emphasis will be placed on the comparison between the Feldkamp-Davis-Kress (FDK) algorithm with the OSC-TV iterative algorithm. Both analytical and iterative reconstruction algorithms will be evaluated considering image fidelity, spatial resolution, detail preservation, noise, artifacts, and how they respond to changes caused by diffusion over time. The FDK algorithm is a well-known analytical cone-beam reconstruction method recognized for its computational efficiency. In contrast, the OSC-TV algorithm combines ordered subsets and total variation regularization to minimize image noise and reconstruction artifacts while maintaining structural details [1].
Although iterative reconstruction methods generally require longer computational times than analytical approaches, they may provide superior image quality by reducing reconstruction artifacts and improving the fidelity of the reconstructed dose distribution.
These improvements can facilitate the visualization of subtle diffusion-induced changes in FXG dosimeters, particularly in low-contrast regions [2]. Image reconstruction used data from experiments with FXG dosimeters.

The reconstructed volumes were analyzed using the VistaDOSE software developed at IPEN/CNEN-SP, for visualization and quantitative analysis of Optical CT reconstructed volumes. The software enables the extraction and comparison of isodose curves, Figure 1 shows reconstructed images and isodose curves. Panels (a,b) correspond to OSC-TV and (c,d) to FDK. Panels (a,c) were generated using the Vista software whereas (b,d) were generated using the VistaDOSE software.
The reconstructed volumes were compared using quantitative image quality metrics and qualitative visual analysis. The OSC-TV algorithm outperformed filtered backprojection (FBP) in terms of image quality by employing a mathematical model that more accurately simulated the physical propagation of light within the object, resulting
in reduced reconstruction artifacts and improved preservation of structural details [1].
The use of iterative reconstruction algorithms provides a more faithful representation of dose distributions while minimizing reconstruction artifacts, supporting the selection of appropriate image reconstruction techniques for three-dimensional FXG dosimetry applications.

Keywords Fricke Gel; reconstruction algorithms; optical CT
Proceedings Publication Intent Yes
Student Award Eligibility yes

Author

Co-authors

Presentation materials