검색
검색 팝업 닫기

Ex) Article Title, Author, Keywords

Article

Split Viewer

Original Article

Progress in Medical Physics 2024; 35(4): 98-105

Published online December 31, 2024

https://doi.org/10.14316/pmp.2024.35.4.98

Copyright © Korean Society of Medical Physics.

Evaluation of Methods for Characterizing Kilovoltage Beam Quality from the Varian TrueBeam STx

Inbum Lee1,2 , Yoonsuk Huh1,2 , Jin Jegal1,2 , Hyojun Park1,2 , Chang Heon Choi1,2,3 , Jung-in Kim1,2,3 , Seonghee Kang1,2,3

1Department of Radiation Oncology, Seoul National University Hospital, Seoul, 2Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, 3Biomedical Research Institute, Seoul National University Hospital, Seoul, Korea

Correspondence to:Seonghee Kang
(kangsh012@gmail.com)
Tel: 82-2-2072-2099
Fax: 82-2-765-3317

Received: October 4, 2024; Revised: December 4, 2024; Accepted: December 10, 2024

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Purpose: This study evaluated various methods for determining the half-value layer (HVL) of kilovoltage (kV) beams produced by the Varian TrueBeam STx on-board imager. By comparing these methods with the standard ionization chamber approach, the study aimed to identify practical solutions for HVL determination and dosimetric characterization of kV beams, particularly in resource-limited settings.
Methods: HVLs for kV beams (40–140 kVp) were measured using an Exradin A12 ionization chamber and a Piranha MULTI meter. The ionization chamber measurements adhered to American Association of Physicists in Medicine Task Group 61 guidelines and served as the reference standard. Additionally, HVL values were calculated using two model-based approaches: SpekPy (a Python-based tool) and Monte Carlo (MC) simulations using Geant4 and GATE. The results from these methods were compared to assess consistency and reliability.
Results: Deviations across all methods were generally below 4%. At 40 kV, the most significant discrepancies were attributed to lower signal levels from the ionization chamber. The consistency between the model-based methods and experimental measurements demonstrates the reliability of these alternative approaches for HVL determination.
Conclusions: Although the ionization chamber remains the gold standard, the Piranha MULTI meter and model-based methods, i.e., SpekPy and MC simulations, have shown promise as viable alternatives, especially in resource-constrained settings. These in silico approaches also offer advantages in convenience and accuracy, supporting their potential for broader future applications.

KeywordsHalf-value-layer, Kilovoltage X-ray beam, Piranha, Monte Carlo simulation, SpekPy

Kilovoltage (kV) imaging systems in modern linear accelerators play a critical role in image-guided radiation therapy by enabling precise patient positioning and accurate target localization [1,2]. The widespread integration of kV imagers in contemporary linear accelerators underscores their essential role in clinical practice.

Despite their benefits, frequent use of kV imaging increases patients’ cumulative radiation dose over treatment courses. Ding and Coffey [3] reported that doses from kV cone beam computed tomography (CT) during a full treatment course can reach up to 300 cGy for radiosensitive organs, with doses to bone being 2–4-fold higher than those to soft tissue, representing a major concerning for pediatric patients. Accurately assessing the dosimetric impact ideally involves using spectrometry to determine the energy spectrum. However, this method demands the use of specialized equipment and meticulous measurements; hence, it is not always available in radiology facilities. A common alternative is determining the half-value layer (HVL) of the X-ray beam. HVL, defined as the thickness of a material (typically aluminum) that reduces beam intensity by 50%, helps derive the linear or mass attenuation coefficient, providing insights into the beam’s dosimetric properties [4,5].

Several methods exist for HVL determination, each with varying requirements for equipment and expertise. The American Association of Physicists in Medicine (AAPM) Task Group 61 (TG-61) protocol for 40–300 kV X-ray beam dosimetry recommends using air-filled ionization chambers with aluminum filters to measure HVL [6]. However, the associated costs and expertise required may limit availability in some clinical settings. Given these challenges, exploring alternative, more accessible HVL determination methods is essential to expand accurate dosimetric assessment capabilities.

This study evaluates multiple approaches for determining HVL using kV beams generated by Varian TrueBeam STx linear accelerator (Varian Medical Systems), a widely used system equipped with a kV on-board imager (OBI) operating at 40–140 kV. Measurements with an ionization chamber and incremental additions of aluminum filters, following AAPM TG-61 guidelines, serve as the reference. The results are compared with HVLs obtained using the Piranha MULTI meter (RTI Group) [7], a simpler, more accessible device with solid-state detectors, as well as calculated values from SpekPy (a Python-based tool derived from SpekCalc) and Monte Carlo (MC) simulations [8,9]. By comparing these methods, the study aims to determine whether alternative approaches can reliably replicate standard ionization chamber results, thereby offering practical solutions for kV beam dosimetric characterization, particularly in resource-limited settings.

The quality of a beam depends on various factors, including the target angle, tube potential, window material, collimator, beam-hardening foil, and bowtie filter [6]. Table 1 lists key components within the TrueBeam’s OBI that influence the quality of its kV beam.

Table 1 TrueBeam OBI’s components in the kV beam path with attenuation equivalents

ItemAttenuation equivalent (mm)
kV tube inherent filtration2.70 Al at 75 kV
Collimator polycarbon window0.30 Al at 100 kV
Beam-hardening foil filter0.89 Ti
Full-fan bowtie filterMinimum 1.53 Al
Maximum 27.42 Al
Half-fan bowtie filterMinimum 1.52 Al
Maximum 27.29 Al

1. Experimental methods

The HVLs of the TrueBeam’s kV OBI, with energies ranging from 40 to 140 kVp, were measured using two detectors: the Piranha MULTI meter and the Exradin A12 ionization chamber (Standard Imaging, Inc.). To maintain consistency, both detectors were positioned at a source-to-detector distance of 100 cm from the kV source. The kV beam was delivered with exposure settings of 200 mA and 200 ms.

The Piranha MULTI meter is a multifunction X-ray measuring device equipped with solid-state detectors for radiography, fluoroscopy, CT, dental applications, and mammography. The factory-calibrated device features “one-shot HVL” capabilities, enabling HVL measurements between 0.72 and 13.00 mm Al, with an accuracy of ±10% or ±0.20 mm for tube voltages of 35–160 kV. After beam delivery, parameters such as tube voltage, exposure time, exposure rate, HVL, and total filtration were recorded in a single exposure.

The Exradin A12 ionization chamber, a waterproof farmer-type chamber with a volume of 0.64 cc, was calibrated at the Korea Research Institute of Standards and Science, National Metrology Institute of Korea, for 60 and 150 kV beams. HVL measurements followed AAPM TG-61 guidelines, ensuring scatter-free and narrow beam conditions. Fig. 1 shows the measurement setup.

Figure 1.Experimental setup, including an Exradin A12 (Standard Imaging, Inc.) ionization chamber, a Cerrobend aperture, and an aluminum filter holder.

The ionization chamber was aligned using the room’s laser system, and the setup was verified using a kV image to confirm the accurate placement of the chamber, aluminum filters, and the aperture (Fig. 2). A 4-cm-diameter Cerrobend aperture was placed between the kV source and ionization chamber, supported by a high-density polyethylene mount at a distance of 50 cm from the chamber, positioning it midway between the source and chamber. Aluminum filters (99.99% purity), ranging from 0.5 mm to several millimeters beyond the HVLs, were incrementally added to measure the relative intensity of transmitted beams. Results were fitted to bi-exponential curves to determine HVL at tube voltages of 40–140 kVp.

Figure 2.Radiographic verification of the setup, showing the ionization chamber placed at the center of the aperture. The image confirms the alignment of the source, aperture, and chamber.

2. Model-based methods

HVL values were computed using SpekPy calculations and MC simulations.

SpekPy energy spectra were calculated using the filtration conditions provided in Table 1, accounting for the TrueBeam system’s inherent 2.70 mm Al filtration, a 14° target angle, a polycarbonate window (0.30 mm Al equivalent), a beam-hardening foil (0.89 mm Ti equivalent), and the full-fan bowtie filter. To simplify geometry, only the minimum thickness value of the full-fan bowtie filter was used, given the small beam area.

The MC model involved beam modeling in Geant4 (version 4.11.3; Geant4 Collaboration) and an experimental setup simulation using GATE (version 9.3; OpenGATE Collaboration). X-ray beam spectra for each tube voltage were first modeled in Geant4, leveraging its multithreading capability. Filtration conditions matched those used in the SpekPy model. However, given that the exact geometry of the TrueBeam’s bowtie filter was unavailable, a CT scan of a Varian Trilogy full-fan bowtie filter was performed using a Brilliance Big Bore CT Scanner (Philips Healthcare). The filter structure was converted to a CAD file and imported for simulation. The minimum and maximum thicknesses of the Trilogy filter were consistent with those of the TrueBeam filter. The energy spectra from SpekPy and Geant4 were compared before GATE modeling.

In GATE, the experimental setup was modeled with a source-to-detector distance of 100 cm. A 4-cm aperture, identical to the Cerrobend aperture, was incorporated to eliminate beam particles outside the aperture. Fig. 3 shows the geometry of the kV source model and the experimental setup model. Simulation results were fitted to bi-exponential functions to estimate HVL for each energy.

Figure 3.Geometry of the modeled components: the kV source (a) and the experimental setup (b). kV, kilovoltage.

The MC-simulated energy spectra closely matched the SpekPy results, verifying the accuracy of the spectral modeling process. Fig. 4 illustrates the overlay of both spectra for each kV setting, showing nearly identical energy distributions across the examined ranges. This validation step reinforced confidence in the MC HVL estimations.

Figure 4.Comparison of energy spectra between SpekPy (SpekCalc) calculations and Geant4 (version 4.11.3; Geant4 Collaboration) simulations. kV, kilovoltage.

Fig. 5 shows the transmission curves derived from the ionization chamber measurements and MC simulation results. These curves depict the attenuation behavior of kV beams through varying aluminum thicknesses. The aluminum thickness required to reduce the relative intensity by half was extracted from each curve to determine the HVLs.

Figure 5.Transmission curves from ionization chamber measurements and GATE (version 9.3; OpenGATE Collaboration) simulations for energies ranging from 40 to 140 kVp, with data fitted to bi-exponential functions. kV, kilovoltage; Al, artificial intelligence.

The HVL values obtained using the ionization chamber, Piranha MULTI meter, and SpekPy calculations are summarized in Fig. 6 and Table 2. The ionization chamber measurements serve as the reference standard, and deviations from other approaches are included in Table 2 to assess consistency across methods.

Table 2 HVL results for each method with tube voltages of 40–140 kVp and deviations from ionization chamber measurements

kVpIonization chamberPiranha (RTI Group)SpekPy (SpekCalc)Monte Carlo




HVL (mm Al)HVL (mm Al)Deviation (%)HVL (mm Al)Deviation (%)HVL (mm Al)Deviation (%)
403.042.70−11.162.68−11.822.62−13.79
604.604.722.534.48−2.684.41−4.21
806.156.160.225.95−3.205.95−3.20
1007.427.490.947.21−2.837.21−2.83
1208.318.552.898.21−1.208.330.25
1409.229.401.969.07−1.629.18−0.43

Figure 6.Plot of HVL results obtained from the four methods. HVL, half-value layer; kV, kilovoltage; Al, artificial intelligence.

For tube voltages of 40–140 kV, the ionization chamber measured HVLs ranging from 3.04 to 9.22 mm Al. The Piranha MULTI meter values spanned 2.70 to 9.40 mm Al, whereas the SpekPy and MC results showed strong agreement, yielding HVLs of 2.62–9.18 mm Al. Across all energies except 140 kV, the SpekPy and MC results were nearly identical, with only minor differences observed at the highest voltage.

The methods demonstrated good overall agreement, with deviations generally below 4%. However, the 40-kV measurements showed marked deviations, attributed primarily to higher values from the ionization chamber, whereas the other three methods were in close agreement. Table 3 presents the relative statistical uncertainties for the ionization chamber measurements and MC simulations.

Table 3 Statistical uncertainties (k=1) of the half-value layers from three sets of ionization chamber measurements and Monte Carlo simulations

kVpIonization chamber (%)Monte Carlo (%)
405.490.23
600.400.81
800.630.31
1000.650.17
1200.050.47
1400.370.22

The results from the various methods demonstrated strong overall agreement, suggesting that both experimental and model-based approaches can reliably characterize the kV beam quality of the TrueBeam OBI. Although ionization chamber measurements remain the gold standard, as recommended by AAPM TG-61, simpler approaches, such as using the Piranha MULTI meter, and computational tools, such as SpekPy and MC simulations, have shown promise as viable alternatives.

In certain scenarios, particularly when rapid kV beam quality assessment is required or conducting experiments is impractical, these alternative methods may be preferred for their convenience. They avoid the need for extensive physical measurements while maintaining accuracy. Determining HVL is not only vital for reference dosimetry but also for characterizing the effective energy of the kV X-ray beam by relating HVL to the linear or mass attenuation coefficient [10]. This underscores the value of reliable HVL estimation, even in the absence of direct experimental measurements.

A notable limitation of this study was the uncertainty in ionization chamber measurements at lower tube voltages, particularly at 40 kVp. At this energy level, the ionization chamber readings dropped to tenths of a picoCoulomb (pC) with aluminum filters in the beam path, resulting in a poor signal-to-noise ratio (SNR). The lowest signal level recorded at 40 kVp was 0.26 pC with 4 mm of aluminum, and noise levels approached 0.1 pC, yielding an SNR of only 2.6. In contrast, at 60 kVp, the SNR exceeded 30, highlighting how low signal levels markedly increased uncertainty at 40 kVp. The statistical uncertainty (k=1) from the three sets of measurements (Table 3) exceeded 5% at 40 kVp, reflecting the impact of the low signal levels.

These discrepancies exceeded standard uncertainty expectations. For example, Aspradakis et al. [11], in Swiss Society of Radiobiology and Medical Physics Recommendation No 9, estimated a relative standard uncertainty from various sources for reference dosimetry with low and medium kV X-rays. The combined uncertainty for low energy, accounting for factors such as reference dosimeter calibration and dose determination in user-specific beams, was 3.22%. Considering only the factors that could have affected our results would have further reduced the expected uncertainty. In our study, uncertainties at other energy levels fell well below 1%, confirming the reliability of our calibration and experimental setup. However, the 5% uncertainty at 40 kVp appears anomalous and may be attributed to low signal levels. Enhancing signal levels could reduce uncertainty and improve the accuracy of HVL determination. Practical approaches include increasing the milliampere-seconds and exposure time to achieve a higher SNR. Alternatively, using a low-energy parallel-plate chamber with a thin entrance window, rather than a cylindrical chamber, could mitigate issues related to low signal levels [12].

The MC simulation results also exhibited slightly higher deviations at lower energies. Although these deviations appeared more pronounced than those from the Piranha MULTI meter or SpekPy, the absolute differences were minimal at 0.07 mm. The relatively small HVL values at lower energy levels magnified these differences. Potential contributing factors include variations in X-ray production efficiency from electron interactions at different energy levels [13], which may not be perfectly captured in simulations.

The HVLs of kV beams produced by the TrueBeam OBI were determined using multiple methods, including experimental measurements, computations, and simulations. The strong agreement across these methods confirms their reliability for characterizing kV beam quality.

This work was supported by Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (20227410100040, Development of patch-type flexible personal dosimeter and real-time remote monitoring system using high-performance inorganic perovskite).

Chang Heon Choi and Jin Jegal are a member of the editorial board of the Progress in Medical Physics, but have no role in the decision to publish this article. The other authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

All relevant data are within the paper and its Supporting Information files.

Conceptualization: Seonghee Kang. Data curation: Inbum Lee, Yoonsuk Huh, Jin Jegal, Hyojun Park, Seonghee Kang. Formal analysis: Seonghee Kang, Chang Heon Choi, Jung-in Kim. Funding acquisition: Seonghee Kang. Investigation: Inbum Lee, Yoonsuk Huh, Jin Jegal, Hyojun Park, Seonghee Kang. Project administration: Seonghee Kang, Chang Heon Choi, Jung-in Kim. Visualization: Inbum Lee, Yoonsuk Huh. Writing – original draft: Inbum Lee, Writing – review & editing: Yoonsuk Huh, Seonghee Kang, Chang Heon Choi, Jung-in Kim.

  1. De Los Santos J, Popple R, Agazaryan N, Bayouth JE, Bissonnette JP, Bucci MK, et al. Image guided radiation therapy (IGRT) technologies for radiation therapy localization and delivery. Int J Radiat Oncol Biol Phys. 2013;87:33-45.
    Pubmed CrossRef
  2. Cho B, Poulsen PR, Keall PJ. Real-time tumor tracking using sequential kV imaging combined with respiratory monitoring: a general framework applicable to commonly used IGRT systems. Phys Med Biol. 2010;55:3299-3316.
    Pubmed KoreaMed CrossRef
  3. Ding GX, Coffey CW. Radiation dose from kilovoltage cone beam computed tomography in an image-guided radiotherapy procedure. Int J Radiat Oncol Biol Phys. 2009;73:610-617.
    Pubmed CrossRef
  4. Seelentag WW, Panzer W. Equivalent half-value thicknesses and mean energies of filtered X-ray bremsstrahlung spectra. Br J Radiol. 1980;53:236-240.
    Pubmed CrossRef
  5. Bushberg J, Anthony Seibert J, Leidholdt EM Jr. , Boone JM, Goldschmidt EJ Jr. The essential physics of medical imaging. Med Phys. 2003;30:1936.
    CrossRef
  6. Ma CM, Coffey CW, DeWerd LA, Liu C, Nath R, Seltzer SM, et al. AAPM protocol for 40-300 kV x-ray beam dosimetry in radiotherapy and radiobiology. Med Phys. 2001;28:868-893.
    Pubmed CrossRef
  7. Lin PP, Goode AR. Accuracy of HVL measurements utilizing solid state detectors for radiography and fluoroscopy X-ray systems. J Appl Clin Med Phys. 2021;22:339-344.
    Pubmed KoreaMed CrossRef
  8. Tran KA, Truong LTH, Mai NV, Dang PN, Vo DTT. Study on the characteristics of X-ray spectra in imaging diagnosis using Monte Carlo simulations. J Korean Phys Soc. 2016;69:1168-1174.
    CrossRef
  9. Ngoc Huy B, Van Dung P. Simulations of X-ray spectra, half value layer, and mean energy from mammography using EGSnrc Monte Carlo and SpekPy. Radiography (Lond). 2023;29:28-37.
    Pubmed CrossRef
  10. Chen SC, Jong WL, Harun AZ. Evaluation of x-ray beam quality based on measurements and estimations using SpekCalc and Ipem78 models. Malays J Med Sci. 2012;19:22-28.
  11. Aspradakis MM, Buchillier T, Kohler G, Kottler C, Krayenbühl J. SSRMP Recommendations No 9: reference dosimetry in low and medium energy x-ray beams. Z Med Phys. 2023;33:601-617.
    Pubmed KoreaMed CrossRef
  12. Li XA, Ma CM, Salhani D. Measurement of percentage depth dose and lateral beam profile for kilovoltage x-ray therapy beams. Phys Med Biol. 1997;42:2561-2568.
    Pubmed CrossRef
  13. Aichinger H, Dierker J, Joite-Barfuß S, Säbel M. Radiation exposure and image quality in x-ray diagnostic radiology. Springer; 2004.
    KoreaMed CrossRef

Article

Original Article

Progress in Medical Physics 2024; 35(4): 98-105

Published online December 31, 2024 https://doi.org/10.14316/pmp.2024.35.4.98

Copyright © Korean Society of Medical Physics.

Evaluation of Methods for Characterizing Kilovoltage Beam Quality from the Varian TrueBeam STx

Inbum Lee1,2 , Yoonsuk Huh1,2 , Jin Jegal1,2 , Hyojun Park1,2 , Chang Heon Choi1,2,3 , Jung-in Kim1,2,3 , Seonghee Kang1,2,3

1Department of Radiation Oncology, Seoul National University Hospital, Seoul, 2Institute of Radiation Medicine, Seoul National University Medical Research Center, Seoul, 3Biomedical Research Institute, Seoul National University Hospital, Seoul, Korea

Correspondence to:Seonghee Kang
(kangsh012@gmail.com)
Tel: 82-2-2072-2099
Fax: 82-2-765-3317

Received: October 4, 2024; Revised: December 4, 2024; Accepted: December 10, 2024

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

Purpose: This study evaluated various methods for determining the half-value layer (HVL) of kilovoltage (kV) beams produced by the Varian TrueBeam STx on-board imager. By comparing these methods with the standard ionization chamber approach, the study aimed to identify practical solutions for HVL determination and dosimetric characterization of kV beams, particularly in resource-limited settings.
Methods: HVLs for kV beams (40–140 kVp) were measured using an Exradin A12 ionization chamber and a Piranha MULTI meter. The ionization chamber measurements adhered to American Association of Physicists in Medicine Task Group 61 guidelines and served as the reference standard. Additionally, HVL values were calculated using two model-based approaches: SpekPy (a Python-based tool) and Monte Carlo (MC) simulations using Geant4 and GATE. The results from these methods were compared to assess consistency and reliability.
Results: Deviations across all methods were generally below 4%. At 40 kV, the most significant discrepancies were attributed to lower signal levels from the ionization chamber. The consistency between the model-based methods and experimental measurements demonstrates the reliability of these alternative approaches for HVL determination.
Conclusions: Although the ionization chamber remains the gold standard, the Piranha MULTI meter and model-based methods, i.e., SpekPy and MC simulations, have shown promise as viable alternatives, especially in resource-constrained settings. These in silico approaches also offer advantages in convenience and accuracy, supporting their potential for broader future applications.

Keywords: Half-value-layer, Kilovoltage X-ray beam, Piranha, Monte Carlo simulation, SpekPy

Introduction

Kilovoltage (kV) imaging systems in modern linear accelerators play a critical role in image-guided radiation therapy by enabling precise patient positioning and accurate target localization [1,2]. The widespread integration of kV imagers in contemporary linear accelerators underscores their essential role in clinical practice.

Despite their benefits, frequent use of kV imaging increases patients’ cumulative radiation dose over treatment courses. Ding and Coffey [3] reported that doses from kV cone beam computed tomography (CT) during a full treatment course can reach up to 300 cGy for radiosensitive organs, with doses to bone being 2–4-fold higher than those to soft tissue, representing a major concerning for pediatric patients. Accurately assessing the dosimetric impact ideally involves using spectrometry to determine the energy spectrum. However, this method demands the use of specialized equipment and meticulous measurements; hence, it is not always available in radiology facilities. A common alternative is determining the half-value layer (HVL) of the X-ray beam. HVL, defined as the thickness of a material (typically aluminum) that reduces beam intensity by 50%, helps derive the linear or mass attenuation coefficient, providing insights into the beam’s dosimetric properties [4,5].

Several methods exist for HVL determination, each with varying requirements for equipment and expertise. The American Association of Physicists in Medicine (AAPM) Task Group 61 (TG-61) protocol for 40–300 kV X-ray beam dosimetry recommends using air-filled ionization chambers with aluminum filters to measure HVL [6]. However, the associated costs and expertise required may limit availability in some clinical settings. Given these challenges, exploring alternative, more accessible HVL determination methods is essential to expand accurate dosimetric assessment capabilities.

This study evaluates multiple approaches for determining HVL using kV beams generated by Varian TrueBeam STx linear accelerator (Varian Medical Systems), a widely used system equipped with a kV on-board imager (OBI) operating at 40–140 kV. Measurements with an ionization chamber and incremental additions of aluminum filters, following AAPM TG-61 guidelines, serve as the reference. The results are compared with HVLs obtained using the Piranha MULTI meter (RTI Group) [7], a simpler, more accessible device with solid-state detectors, as well as calculated values from SpekPy (a Python-based tool derived from SpekCalc) and Monte Carlo (MC) simulations [8,9]. By comparing these methods, the study aims to determine whether alternative approaches can reliably replicate standard ionization chamber results, thereby offering practical solutions for kV beam dosimetric characterization, particularly in resource-limited settings.

Materials and Methods

The quality of a beam depends on various factors, including the target angle, tube potential, window material, collimator, beam-hardening foil, and bowtie filter [6]. Table 1 lists key components within the TrueBeam’s OBI that influence the quality of its kV beam.

Table 1 . TrueBeam OBI’s components in the kV beam path with attenuation equivalents.

ItemAttenuation equivalent (mm)
kV tube inherent filtration2.70 Al at 75 kV
Collimator polycarbon window0.30 Al at 100 kV
Beam-hardening foil filter0.89 Ti
Full-fan bowtie filterMinimum 1.53 Al
Maximum 27.42 Al
Half-fan bowtie filterMinimum 1.52 Al
Maximum 27.29 Al


1. Experimental methods

The HVLs of the TrueBeam’s kV OBI, with energies ranging from 40 to 140 kVp, were measured using two detectors: the Piranha MULTI meter and the Exradin A12 ionization chamber (Standard Imaging, Inc.). To maintain consistency, both detectors were positioned at a source-to-detector distance of 100 cm from the kV source. The kV beam was delivered with exposure settings of 200 mA and 200 ms.

The Piranha MULTI meter is a multifunction X-ray measuring device equipped with solid-state detectors for radiography, fluoroscopy, CT, dental applications, and mammography. The factory-calibrated device features “one-shot HVL” capabilities, enabling HVL measurements between 0.72 and 13.00 mm Al, with an accuracy of ±10% or ±0.20 mm for tube voltages of 35–160 kV. After beam delivery, parameters such as tube voltage, exposure time, exposure rate, HVL, and total filtration were recorded in a single exposure.

The Exradin A12 ionization chamber, a waterproof farmer-type chamber with a volume of 0.64 cc, was calibrated at the Korea Research Institute of Standards and Science, National Metrology Institute of Korea, for 60 and 150 kV beams. HVL measurements followed AAPM TG-61 guidelines, ensuring scatter-free and narrow beam conditions. Fig. 1 shows the measurement setup.

Figure 1. Experimental setup, including an Exradin A12 (Standard Imaging, Inc.) ionization chamber, a Cerrobend aperture, and an aluminum filter holder.

The ionization chamber was aligned using the room’s laser system, and the setup was verified using a kV image to confirm the accurate placement of the chamber, aluminum filters, and the aperture (Fig. 2). A 4-cm-diameter Cerrobend aperture was placed between the kV source and ionization chamber, supported by a high-density polyethylene mount at a distance of 50 cm from the chamber, positioning it midway between the source and chamber. Aluminum filters (99.99% purity), ranging from 0.5 mm to several millimeters beyond the HVLs, were incrementally added to measure the relative intensity of transmitted beams. Results were fitted to bi-exponential curves to determine HVL at tube voltages of 40–140 kVp.

Figure 2. Radiographic verification of the setup, showing the ionization chamber placed at the center of the aperture. The image confirms the alignment of the source, aperture, and chamber.

2. Model-based methods

HVL values were computed using SpekPy calculations and MC simulations.

SpekPy energy spectra were calculated using the filtration conditions provided in Table 1, accounting for the TrueBeam system’s inherent 2.70 mm Al filtration, a 14° target angle, a polycarbonate window (0.30 mm Al equivalent), a beam-hardening foil (0.89 mm Ti equivalent), and the full-fan bowtie filter. To simplify geometry, only the minimum thickness value of the full-fan bowtie filter was used, given the small beam area.

The MC model involved beam modeling in Geant4 (version 4.11.3; Geant4 Collaboration) and an experimental setup simulation using GATE (version 9.3; OpenGATE Collaboration). X-ray beam spectra for each tube voltage were first modeled in Geant4, leveraging its multithreading capability. Filtration conditions matched those used in the SpekPy model. However, given that the exact geometry of the TrueBeam’s bowtie filter was unavailable, a CT scan of a Varian Trilogy full-fan bowtie filter was performed using a Brilliance Big Bore CT Scanner (Philips Healthcare). The filter structure was converted to a CAD file and imported for simulation. The minimum and maximum thicknesses of the Trilogy filter were consistent with those of the TrueBeam filter. The energy spectra from SpekPy and Geant4 were compared before GATE modeling.

In GATE, the experimental setup was modeled with a source-to-detector distance of 100 cm. A 4-cm aperture, identical to the Cerrobend aperture, was incorporated to eliminate beam particles outside the aperture. Fig. 3 shows the geometry of the kV source model and the experimental setup model. Simulation results were fitted to bi-exponential functions to estimate HVL for each energy.

Figure 3. Geometry of the modeled components: the kV source (a) and the experimental setup (b). kV, kilovoltage.

Results

The MC-simulated energy spectra closely matched the SpekPy results, verifying the accuracy of the spectral modeling process. Fig. 4 illustrates the overlay of both spectra for each kV setting, showing nearly identical energy distributions across the examined ranges. This validation step reinforced confidence in the MC HVL estimations.

Figure 4. Comparison of energy spectra between SpekPy (SpekCalc) calculations and Geant4 (version 4.11.3; Geant4 Collaboration) simulations. kV, kilovoltage.

Fig. 5 shows the transmission curves derived from the ionization chamber measurements and MC simulation results. These curves depict the attenuation behavior of kV beams through varying aluminum thicknesses. The aluminum thickness required to reduce the relative intensity by half was extracted from each curve to determine the HVLs.

Figure 5. Transmission curves from ionization chamber measurements and GATE (version 9.3; OpenGATE Collaboration) simulations for energies ranging from 40 to 140 kVp, with data fitted to bi-exponential functions. kV, kilovoltage; Al, artificial intelligence.

The HVL values obtained using the ionization chamber, Piranha MULTI meter, and SpekPy calculations are summarized in Fig. 6 and Table 2. The ionization chamber measurements serve as the reference standard, and deviations from other approaches are included in Table 2 to assess consistency across methods.

Table 2 . HVL results for each method with tube voltages of 40–140 kVp and deviations from ionization chamber measurements.

kVpIonization chamberPiranha (RTI Group)SpekPy (SpekCalc)Monte Carlo




HVL (mm Al)HVL (mm Al)Deviation (%)HVL (mm Al)Deviation (%)HVL (mm Al)Deviation (%)
403.042.70−11.162.68−11.822.62−13.79
604.604.722.534.48−2.684.41−4.21
806.156.160.225.95−3.205.95−3.20
1007.427.490.947.21−2.837.21−2.83
1208.318.552.898.21−1.208.330.25
1409.229.401.969.07−1.629.18−0.43


Figure 6. Plot of HVL results obtained from the four methods. HVL, half-value layer; kV, kilovoltage; Al, artificial intelligence.

For tube voltages of 40–140 kV, the ionization chamber measured HVLs ranging from 3.04 to 9.22 mm Al. The Piranha MULTI meter values spanned 2.70 to 9.40 mm Al, whereas the SpekPy and MC results showed strong agreement, yielding HVLs of 2.62–9.18 mm Al. Across all energies except 140 kV, the SpekPy and MC results were nearly identical, with only minor differences observed at the highest voltage.

The methods demonstrated good overall agreement, with deviations generally below 4%. However, the 40-kV measurements showed marked deviations, attributed primarily to higher values from the ionization chamber, whereas the other three methods were in close agreement. Table 3 presents the relative statistical uncertainties for the ionization chamber measurements and MC simulations.

Table 3 . Statistical uncertainties (k=1) of the half-value layers from three sets of ionization chamber measurements and Monte Carlo simulations.

kVpIonization chamber (%)Monte Carlo (%)
405.490.23
600.400.81
800.630.31
1000.650.17
1200.050.47
1400.370.22

Discussion

The results from the various methods demonstrated strong overall agreement, suggesting that both experimental and model-based approaches can reliably characterize the kV beam quality of the TrueBeam OBI. Although ionization chamber measurements remain the gold standard, as recommended by AAPM TG-61, simpler approaches, such as using the Piranha MULTI meter, and computational tools, such as SpekPy and MC simulations, have shown promise as viable alternatives.

In certain scenarios, particularly when rapid kV beam quality assessment is required or conducting experiments is impractical, these alternative methods may be preferred for their convenience. They avoid the need for extensive physical measurements while maintaining accuracy. Determining HVL is not only vital for reference dosimetry but also for characterizing the effective energy of the kV X-ray beam by relating HVL to the linear or mass attenuation coefficient [10]. This underscores the value of reliable HVL estimation, even in the absence of direct experimental measurements.

A notable limitation of this study was the uncertainty in ionization chamber measurements at lower tube voltages, particularly at 40 kVp. At this energy level, the ionization chamber readings dropped to tenths of a picoCoulomb (pC) with aluminum filters in the beam path, resulting in a poor signal-to-noise ratio (SNR). The lowest signal level recorded at 40 kVp was 0.26 pC with 4 mm of aluminum, and noise levels approached 0.1 pC, yielding an SNR of only 2.6. In contrast, at 60 kVp, the SNR exceeded 30, highlighting how low signal levels markedly increased uncertainty at 40 kVp. The statistical uncertainty (k=1) from the three sets of measurements (Table 3) exceeded 5% at 40 kVp, reflecting the impact of the low signal levels.

These discrepancies exceeded standard uncertainty expectations. For example, Aspradakis et al. [11], in Swiss Society of Radiobiology and Medical Physics Recommendation No 9, estimated a relative standard uncertainty from various sources for reference dosimetry with low and medium kV X-rays. The combined uncertainty for low energy, accounting for factors such as reference dosimeter calibration and dose determination in user-specific beams, was 3.22%. Considering only the factors that could have affected our results would have further reduced the expected uncertainty. In our study, uncertainties at other energy levels fell well below 1%, confirming the reliability of our calibration and experimental setup. However, the 5% uncertainty at 40 kVp appears anomalous and may be attributed to low signal levels. Enhancing signal levels could reduce uncertainty and improve the accuracy of HVL determination. Practical approaches include increasing the milliampere-seconds and exposure time to achieve a higher SNR. Alternatively, using a low-energy parallel-plate chamber with a thin entrance window, rather than a cylindrical chamber, could mitigate issues related to low signal levels [12].

The MC simulation results also exhibited slightly higher deviations at lower energies. Although these deviations appeared more pronounced than those from the Piranha MULTI meter or SpekPy, the absolute differences were minimal at 0.07 mm. The relatively small HVL values at lower energy levels magnified these differences. Potential contributing factors include variations in X-ray production efficiency from electron interactions at different energy levels [13], which may not be perfectly captured in simulations.

Conclusions

The HVLs of kV beams produced by the TrueBeam OBI were determined using multiple methods, including experimental measurements, computations, and simulations. The strong agreement across these methods confirms their reliability for characterizing kV beam quality.

Funding

This work was supported by Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government (MOTIE) (20227410100040, Development of patch-type flexible personal dosimeter and real-time remote monitoring system using high-performance inorganic perovskite).

Conflicts of Interest

Chang Heon Choi and Jin Jegal are a member of the editorial board of the Progress in Medical Physics, but have no role in the decision to publish this article. The other authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Availability of Data and Materials

All relevant data are within the paper and its Supporting Information files.

Author Contributions

Conceptualization: Seonghee Kang. Data curation: Inbum Lee, Yoonsuk Huh, Jin Jegal, Hyojun Park, Seonghee Kang. Formal analysis: Seonghee Kang, Chang Heon Choi, Jung-in Kim. Funding acquisition: Seonghee Kang. Investigation: Inbum Lee, Yoonsuk Huh, Jin Jegal, Hyojun Park, Seonghee Kang. Project administration: Seonghee Kang, Chang Heon Choi, Jung-in Kim. Visualization: Inbum Lee, Yoonsuk Huh. Writing – original draft: Inbum Lee, Writing – review & editing: Yoonsuk Huh, Seonghee Kang, Chang Heon Choi, Jung-in Kim.

Fig 1.

Figure 1.Experimental setup, including an Exradin A12 (Standard Imaging, Inc.) ionization chamber, a Cerrobend aperture, and an aluminum filter holder.
Progress in Medical Physics 2024; 35: 98-105https://doi.org/10.14316/pmp.2024.35.4.98

Fig 2.

Figure 2.Radiographic verification of the setup, showing the ionization chamber placed at the center of the aperture. The image confirms the alignment of the source, aperture, and chamber.
Progress in Medical Physics 2024; 35: 98-105https://doi.org/10.14316/pmp.2024.35.4.98

Fig 3.

Figure 3.Geometry of the modeled components: the kV source (a) and the experimental setup (b). kV, kilovoltage.
Progress in Medical Physics 2024; 35: 98-105https://doi.org/10.14316/pmp.2024.35.4.98

Fig 4.

Figure 4.Comparison of energy spectra between SpekPy (SpekCalc) calculations and Geant4 (version 4.11.3; Geant4 Collaboration) simulations. kV, kilovoltage.
Progress in Medical Physics 2024; 35: 98-105https://doi.org/10.14316/pmp.2024.35.4.98

Fig 5.

Figure 5.Transmission curves from ionization chamber measurements and GATE (version 9.3; OpenGATE Collaboration) simulations for energies ranging from 40 to 140 kVp, with data fitted to bi-exponential functions. kV, kilovoltage; Al, artificial intelligence.
Progress in Medical Physics 2024; 35: 98-105https://doi.org/10.14316/pmp.2024.35.4.98

Fig 6.

Figure 6.Plot of HVL results obtained from the four methods. HVL, half-value layer; kV, kilovoltage; Al, artificial intelligence.
Progress in Medical Physics 2024; 35: 98-105https://doi.org/10.14316/pmp.2024.35.4.98

Table 1 TrueBeam OBI’s components in the kV beam path with attenuation equivalents

ItemAttenuation equivalent (mm)
kV tube inherent filtration2.70 Al at 75 kV
Collimator polycarbon window0.30 Al at 100 kV
Beam-hardening foil filter0.89 Ti
Full-fan bowtie filterMinimum 1.53 Al
Maximum 27.42 Al
Half-fan bowtie filterMinimum 1.52 Al
Maximum 27.29 Al

Table 2 HVL results for each method with tube voltages of 40–140 kVp and deviations from ionization chamber measurements

kVpIonization chamberPiranha (RTI Group)SpekPy (SpekCalc)Monte Carlo




HVL (mm Al)HVL (mm Al)Deviation (%)HVL (mm Al)Deviation (%)HVL (mm Al)Deviation (%)
403.042.70−11.162.68−11.822.62−13.79
604.604.722.534.48−2.684.41−4.21
806.156.160.225.95−3.205.95−3.20
1007.427.490.947.21−2.837.21−2.83
1208.318.552.898.21−1.208.330.25
1409.229.401.969.07−1.629.18−0.43

Table 3 Statistical uncertainties (k=1) of the half-value layers from three sets of ionization chamber measurements and Monte Carlo simulations

kVpIonization chamber (%)Monte Carlo (%)
405.490.23
600.400.81
800.630.31
1000.650.17
1200.050.47
1400.370.22

References

  1. De Los Santos J, Popple R, Agazaryan N, Bayouth JE, Bissonnette JP, Bucci MK, et al. Image guided radiation therapy (IGRT) technologies for radiation therapy localization and delivery. Int J Radiat Oncol Biol Phys. 2013;87:33-45.
    Pubmed CrossRef
  2. Cho B, Poulsen PR, Keall PJ. Real-time tumor tracking using sequential kV imaging combined with respiratory monitoring: a general framework applicable to commonly used IGRT systems. Phys Med Biol. 2010;55:3299-3316.
    Pubmed KoreaMed CrossRef
  3. Ding GX, Coffey CW. Radiation dose from kilovoltage cone beam computed tomography in an image-guided radiotherapy procedure. Int J Radiat Oncol Biol Phys. 2009;73:610-617.
    Pubmed CrossRef
  4. Seelentag WW, Panzer W. Equivalent half-value thicknesses and mean energies of filtered X-ray bremsstrahlung spectra. Br J Radiol. 1980;53:236-240.
    Pubmed CrossRef
  5. Bushberg J, Anthony Seibert J, Leidholdt EM Jr. , Boone JM, Goldschmidt EJ Jr. The essential physics of medical imaging. Med Phys. 2003;30:1936.
    CrossRef
  6. Ma CM, Coffey CW, DeWerd LA, Liu C, Nath R, Seltzer SM, et al. AAPM protocol for 40-300 kV x-ray beam dosimetry in radiotherapy and radiobiology. Med Phys. 2001;28:868-893.
    Pubmed CrossRef
  7. Lin PP, Goode AR. Accuracy of HVL measurements utilizing solid state detectors for radiography and fluoroscopy X-ray systems. J Appl Clin Med Phys. 2021;22:339-344.
    Pubmed KoreaMed CrossRef
  8. Tran KA, Truong LTH, Mai NV, Dang PN, Vo DTT. Study on the characteristics of X-ray spectra in imaging diagnosis using Monte Carlo simulations. J Korean Phys Soc. 2016;69:1168-1174.
    CrossRef
  9. Ngoc Huy B, Van Dung P. Simulations of X-ray spectra, half value layer, and mean energy from mammography using EGSnrc Monte Carlo and SpekPy. Radiography (Lond). 2023;29:28-37.
    Pubmed CrossRef
  10. Chen SC, Jong WL, Harun AZ. Evaluation of x-ray beam quality based on measurements and estimations using SpekCalc and Ipem78 models. Malays J Med Sci. 2012;19:22-28.
  11. Aspradakis MM, Buchillier T, Kohler G, Kottler C, Krayenbühl J. SSRMP Recommendations No 9: reference dosimetry in low and medium energy x-ray beams. Z Med Phys. 2023;33:601-617.
    Pubmed KoreaMed CrossRef
  12. Li XA, Ma CM, Salhani D. Measurement of percentage depth dose and lateral beam profile for kilovoltage x-ray therapy beams. Phys Med Biol. 1997;42:2561-2568.
    Pubmed CrossRef
  13. Aichinger H, Dierker J, Joite-Barfuß S, Säbel M. Radiation exposure and image quality in x-ray diagnostic radiology. Springer; 2004.
    KoreaMed CrossRef
Korean Society of Medical Physics

Vol.35 No.4
December 2024

pISSN 2508-4445
eISSN 2508-4453
Formerly ISSN 1226-5829

Frequency: Quarterly

Current Issue   |   Archives

Stats or Metrics

Share this article on :

  • line