Comparison of different solutions for applying patient-specific beam quality in treatment planning CT imaging.
5 affiliations
- 1Oncologic Imaging and Invasive Diagnostic Centre, National Institute of Oncology, Budapest, Hungary.
- 2Institute of Nuclear Techniques, Budapest University of Technology and Economics, Budapest, Hungary.
- 3Center of Radiotherapy, National Institute of Oncology, Budapest, Hungary.
- 4Department of Oncology, Semmelweis University, Budapest, Hungary.
- 5National Tumor Biology Laboratory, National Institute of Oncology, Budapest, Hungary.
Abstract
For radiotherapy treatment planning, it is necessary to convert Hounsfield units from CT images to electron density. This conversion can be performed using separate calibration curves for each beam quality. Another available solution is to use an Artificial 120 kVp reconstruction, allowing the use of a single calibration curve. The Direct Density kernel is a more modern approach that directly reconstructs electron density and requires no calibration measurements. The Gammex RMI 467 phantom was measured at different positions: the central axis and an anterior offset from it. The patient-size dependence was tested by measuring the Gammex inserts in two in-house-made phantoms with variable water-equivalent diameters. It was found that the standard reconstruction with separate calibration curves is the most accurate solution. The Artificial 120 kVp reconstruction introduces slightly greater uncertainty in the reconstructed electron density (ED), increasing the tube potential dependence of the reconstructed ED. The Direct Density reconstruction is less accurate; however, its performance can be improved by calibration. A traditional kernel with calibration curves for different energies results in more accurate electron density determination than the Artificial 120 kVp or the Direct Density reconstruction. One of the most widely used treatment planning systems has recently introduced the option to add multiple calibration curves for the same CT model.
Related
- Comparison of Different Models in Calculation of Tumor Control Probability and Normal Tissue Complication Probability in Breast Cancer Radiotherapy.
- Comparison of different radiotherapy planning techniques in breast cancer using active breathing coordinator
- Dosimetric and Treatment Delivery Efficiency Comparison of Two Ring-gantry Linear Accelerator Platforms for Liver Stereotactic Body Radiotherapy
- Consistent treatment of flavour-specific neutrino self-interactions in cosmology
- Treatment plan quality comparison between ZAP-Axon® Krylov optimizer and Leksell Gamma Knife® Lightning optimizer for brain metastasis.
BibTeX
@article{sudar2026comparison,
title = {{Comparison of different solutions for applying patient-specific beam quality in treatment planning CT imaging.}},
author = {Ákos Sudár and Tamás Pócza and Domonkos Szegedi and Tibor Major and Ádám Domonkos Tárnoki and Dávid László Tárnoki and Csilla Pesznyák},
journal = {Reports of practical oncology and radiotherapy : journal of Greatpoland Cancer Center in Poznan and Polish Society of Radiation Oncology},
year = {2026},
month = nov,
doi = {10.5603/rpor.113351},
url = {https://europepmc.org/article/MED/42997854},
note = {CC BY-NC-ND},
}
RIS
TY - JOUR
TI - Comparison of different solutions for applying patient-specific beam quality in treatment planning CT imaging.
AU - Ákos Sudár
AU - Tamás Pócza
AU - Domonkos Szegedi
AU - Tibor Major
AU - Ádám Domonkos Tárnoki
AU - Dávid László Tárnoki
AU - Csilla Pesznyák
PY - 2026
DA - 2026/11/01
JO - Reports of practical oncology and radiotherapy : journal of Greatpoland Cancer Center in Poznan and Polish Society of Radiation Oncology
DO - 10.5603/rpor.113351
UR - https://europepmc.org/article/MED/42997854
AB - For radiotherapy treatment planning, it is necessary to convert Hounsfield units from CT images to electron density. This conversion can be performed using separate calibration curves for each beam quality. Another available solution is to use an Artificial 120 kVp reconstruction, allowing the use of a single calibration curve. The Direct Density kernel is a more modern approach that directly reconstructs electron density and requires no calibration measurements. The Gammex RMI 467 phantom was measured at different positions: the central axis and an anterior offset from it. The patient-size dependence was tested by measuring the Gammex inserts in two in-house-made phantoms with variable water-equivalent diameters. It was found that the standard reconstruction with separate calibration curves is the most accurate solution. The Artificial 120 kVp reconstruction introduces slightly greater uncertainty in the reconstructed electron density (ED), increasing the tube potential dependence of the reconstructed ED. The Direct Density reconstruction is less accurate; however, its performance can be improved by calibration. A traditional kernel with calibration curves for different energies results in more accurate electron density determination than the Artificial 120 kVp or the Direct Density reconstruction. One of the most widely used treatment planning systems has recently introduced the option to add multiple calibration curves for the same CT model.
SN - 1507-1367
SN - 2083-4640
ER -