Proton Beam Range and Charge Verification Using Multilayer Faraday Collector

Abstract

Purpose

A daily quality assurance (QA) check in proton therapy is ensuring that the range of each proton beam energy in water is accurate to 1 mm. This is important for ensuring that the tumor is adequately irradiated while minimizing damage to surrounding healthy tissue. It is also important to verify the total charge collected against the beam model. This work proposes a time-efficient method for verifying the range and total charge of proton beams at different energies using a multilayer Faraday collector (MLFC).

Methods

We used an MLFC-128-250 MeV comprising 128 layers of thin copper foils separated by thin insulating Kapton^TM layers. Protons passing through the collector induce a charge on the metallic foils, which is integrated and measured by a multichannel electrometer. The charge deposition on the foils provides information about the beam range.

Results

Our results show that the proton beam range obtained using MLFC correlates closely with the range obtained from commissioning water tank measurements for all proton energies. Upon applying a range calibration factor, the maximum deviation is 0.4 g/cm². The MLFC range showed no dependence on the number of monitor units and the source-to-surface distance. Range measurements collected over multiple weeks exhibited stability. The total charge collected agrees closely with the theoretical charge from the treatment planning system beam model for low- and mid-range energies.

Conclusions

We have calibrated and commissioned the use of the MLFC to easily verify range and total charge of proton beams. This tool will improve the workflow efficiency of the proton QA.

Keywords

proton beam therapy quality assurance multi-layer Faraday collector proton range

Introduction

Proton beam therapy (PBT) is a rapidly growing technique in cancer treatment due to its precision and low side effects.^1,2 The key advantage of PBT lies in its ability to deliver a highly conformal dose to the target, hence giving a superior spatial dose distribution in the patient.^1,2 With increasing interest and adoption of PBT globally,³ it is crucial to develop quality assurance (QA) procedures and methods that ensure the safe delivery of treatment to patients.

As recommended by the American Association of Physicists in Medicine (AAPM) Task Group 224 report,⁴ one of the key beam parameters to be checked daily is the range of the pristine Bragg peak in water, which is dependent on the beam energy. The International Commission on Radiation Units and Measurements (ICRU) 78 report⁵ defines the beam range in water as the depth at the distal 90% dose level (ie d_d90). For a pencil beam scanning (PBS) system, the daily, monthly, and annual QA tolerances are recommended to be within ±1 mm from baseline.

The gold standard for measuring depth doses typically involves the use of a scanning water tank and ionization chamber, which requires some training to operate. As the scanning field is to be delivered over the entire volume numerous times, the process is time-consuming and could take hours to measure the depth-dose curve for all energies. To improve the ease and efficiency of daily range verification, commercial products such as the multilayer ionization chamber^6,7 introduced by IBA (IBA Dosimetry) and PTW, as well as the Ranger-300 (Logos Systems),⁸ have been introduced to reduce the operational time of range measurements.

The multilayer Faraday collector (MLFC) was first designed by Gottschalk to verify proton beam range at the Northeast Proton Therapy Center.⁹ Tesfamicael et al¹⁰ concluded that the MLFC can be used to perform a quick and precise daily range verification of proton beams and reported a submillimetre daily range variation. Boisseau et al¹¹ similarly reported good reproducibility of range measurements using the MLFC and verified the measured beam spectra with TOol for PArticle Simulation (TOPAS) calculations. Jung et al¹² compared 4 measured energies using an MLFC with energies calculated using Stopping and Range of Ions in Matter simulations and reported good agreement within 1%.

In this work, we evaluate the use of the MLFC as a QA tool for range verification at the National Cancer Centre Singapore. We test the MLFC for range accuracy, reproducibility, and range dependence on the source-to-surface distance (SSD) and number of monitor units (MU). Apart from range verification, we also present a method to measure number of protons using the MLFC, which is a parameter used in our treatment planning system (TPS) for Monte Carlo simulation. Testa et al¹³ have previously employed a comprehensive multistep approach to validate MLFC charge measurements. They utilized TOPAS simulations, validated with experimental output factor measurements involving ion chamber charge collection in water tank setups, to assess the consistency and reliability of MLFC charge measurements. The validation involved multiple steps, which provided valuable insights into MLFC charge collection consistency. In this work, the measured proton charge on the MLFC was directly validated with absolute dosimetry data collected empirically using a water tank and ionization chamber set-up, which has not been reported previously to our knowledge. We believe the methodology in this work will be relevant to many centers looking for efficient ways to verify the beam range and charge of their beamlines.

Methods

MLFC set-up

In this study, the MLFC-128-250 MeV (Pyramid Technical Consultants) was used.¹⁴ It is a compact device measuring approximately 120 by 120 by 145 mm and weighing 5.6 kg. The MLFC consists of 128 layers of thin pure copper foils of 0.508 mm thick, separated by thin insulating Kapton layers of 0.05 mm thick. The densities of copper and Kapton are taken to be 8.96 and 1.42 g/cm³ respectively in this study. The entrance has a sensitive area of 72 mm in diameter. The protons passing through the MLFC induce a charge on the metallic foils, which is collected and measured by a 128-channel electrometer. The charge deposition on the foils provides information about the range of the proton beam. The strip readout electronics comprises 2 Centronics VHDCI (Very High Density Cable Interconnect) 68-way receptacle with gold-plated contacts. The amount of charge collected in each channel across time, as well as the mean peak channel (ie position of foil with the highest charge collection), is reported by and stored in the Windows host software application included with the MLFC-128. The MLFC collects measurements at 50-millisecond intervals, and we used the maximum integration time, $T_{mea}$ , of 1 millisecond in this work, hence there is a deadtime, $T_{dead}$ of 49 milliseconds where no measurements are taken.

At the National Cancer Centre Singapore, a Hitachi ProBeat proton therapy system (Hitachi Ltd) with PBS and 98 discrete energy layers ranging from 70.2 to 228.7 MeV is used. The Hitachi ProBeat is a synchrotron-based system with a pulsed or bunched beam structure. The protons are extracted in random pulses using the slow resonant extraction technique. The radiofrequency is about 1 kHz.¹⁵ The operation cycle is about 4 s for the highest energy and 12 MUs are extracted. The ProBeat also has a constant dose rate of 8 MU/s. In our institution, 1 MU gives approximately 4×10⁸ protons at the lowest energy of 70.2 MeV and 9×10⁸ protons at the highest energy of 228.7 MeV.

The MLFC was placed downstream of the nozzle and aligned to the room lasers’ isocenter, requiring a set-up time of under 5 min. Figure 1A shows a schematic of the MLFC and some of its copper and Kapton layers, showing how protons stopping at a copper foil are collected by the electrometer. If the protons stopped in a Kapton layer, it induces a mirror charge in a neighboring copper layer, hence the charge will still be counted. Figure 1C shows the set-up of the MLFC in the room. For each measurement, 100 MU of single-spot proton beams were delivered toward the MLFC using 10 initial beamline energies ranging from 70.2 to 228.7 MeV. The mean peak channel as well as the total charge were recorded for each measurement. To check for reproducibility, 3 measurements per energy were obtained every 2 weeks over a span of 11 weeks.

Figure 1.

(A) Schematic of MLFC and some of its copper and Kapton layers, showing how protons stopping at a copper foil are collected by the electrometer. The blue solid line depicts the proton Bragg peak, while the red dotted line depicts the relative signal collected by the layer. (B) Total stopping power of protons in water, copper, and Kapton obtained from PSTAR, NIST. (C) Experimental set-up of MLFC at NCCS. The Orfit indexing bar (Orfit Industries) ensures that the MLFC is parallel to the nozzle, which is parked at 90°. The MLFC is then connected to a 128-channel electrometer (not shown in picture). Abbreviations: MLFC, multilayer Faraday collector; NIST, National Institute of Standards and Technology.

Range Calculation

At NCCS, we follow the ICRU 78 report's definition of range as d_d90. It is not clear if the peak channel measurements, which is the result of protons that were completely stopped in the MLFC layers, should correspond to d_d90, d_d80, or the depth at maximum dose (ie d_max), and more studies will be required to investigate this. Nonetheless, we verified the linearity of the peak channels obtained from MLFC measurements across the range of treatment energies with d_d90, d_d80, and d_max values measured using a PTW MP3 water tank¹⁶ and PTW 34089 ionization chamber¹⁷ setup during gantry commissioning 3 months earlier.

The MLFC is not designed to directly measure depth-dose curves in water.¹¹ Instead, its advantage lies in the direct linear relationship between its peak channel measurements and the positions of the Bragg peak. However, to check for device integrity, we calculated the theoretical range of the proton beam in water using the peak channel measurements in this study. To calculate the proton range from the peak channel, the water equivalent thickness, $t_{w}$ , of each copper and Kapton layer was determined using Equation (1)¹⁸:

\begin{matrix} t_{w} = t_{m} \frac{ρ_{m}}{ρ_{w}} \frac{{\bar{S}}_{m}}{{\bar{S}}_{w}} \end{matrix}

(1)

where

t_{m}

is the physical thickness of the material,

ρ_{m}

and

ρ_{w}

are the mass densities of the material and water respectively, and

{\bar{S}}_{m}

and

{\bar{S}}_{w}

are the mean proton mass stopping power values of the material and water respectively, given by Equation (2):

\begin{matrix} \bar{S} = \frac{\int_{E} S d E}{\int_{E} d E} \end{matrix}

(2)

The stopping power of protons at different energies for the materials were obtained from the PSTAR program by the National Institute of Standards and Technology (NIST).¹⁹ Figure 1B shows the total stopping power of protons of different energies in water, copper, and Kapton.

By multiplying the peak channel with the $t_{w}$ of the corresponding number of copper and Kapton layers, the range of the proton beams in water can be derived. This was then compared with the proton beam range measured using the water tank setup. A range calibration factor was subsequently applied to bring the range values derived from the MLFC peak channel measurements closer to the water tank range values.

This study also investigated the dependence of the derived MLFC range on the number of MU used and the SSD.

Charge Determination

The advantage of the MLFC lies in its ability to also measure charge, which is useful for validation of the beam model on the TPS. There are 2 conventional ways to calibrate a beam monitor.^20,21 The first method is to derive the dose-area-product to water from the absorbed dose to water with a plane-parallel ionization chamber in single-energy scanned beams. The second method is to determine the number of particles delivered by a narrow mono-energetic beam for each spot using a large-area plane-parallel ionization chamber.

In this work, we determined the total charge per MU, $Q_{total}$ , collected by the MLFC by summing the collected charge, $Q_{coll}$ , per MU across all channels and over the irradiation duration. Assuming a constant dose rate, $Q_{total}$ is then obtained by multiplying $Q_{coll}$ per MU by a factor equal to the ratio of measurement interval time to the measurement integration time, as shown in Equation (3):

\begin{matrix} Q_{total} = \sum_{T_{mea}} \sum_{channel} \frac{Q_{coll, channel}}{no . of MUs} \times \frac{T_{mea} + T_{dead}}{T_{mea}} \end{matrix}

(3)

The number of protons,

n_{p}

, can be derived by dividing

Q_{total}

by the proton charge, as shown in Equation (4):

\begin{matrix} n_{p} = \frac{Q_{total}}{1.6 \times 10^{- 19} C} \end{matrix}

(4)

The measured

n_{p}

was then compared with the ion count on the RayStation 10B (RaySearch Laboratories) TPS, which was calculated using the absolute dosimetry data collected using the water tank and the Advanced Markus chamber. A monoenergetic rectilinear spot patterns of 10 × 10 cm² and spot spacing of 2.5 mm were irradiated on the ionization chamber placed at 2 cm water depth. The absolute dose measured in this manner can be related to the

n_{p}

based on a simple scaling factor by the CEMA (converted energy per unit mass).

The reproducibility of total charge collected by the MLFC was also evaluated using the measurements taken fortnightly across 11 weeks.

Results

Range

The charge collected across the 128 channels of the MLFC follows a Gaussian distribution, as seen by an example of a 228.7 MeV proton beam in Figure 2A. The mean peak channels obtained from MLFC measurements for the 10 energies correlate strongly with d_d90, d_d80, and d_max, giving Pearson's correlation coefficients of 1.0, as seen in Figure 2B.

Figure 2.

(A) Integral charge for 100 MU collected over 128 channels for a 228.7 Mev single-spot proton beam, with Gaussian fitting (in red). (B) Plot of MLFC peak channel with the d_d90, d_d80, and d_max values measured using a water tank for 10 proton energies. Abbreviations: MLFC, multilayer Faraday collector; MU, monitor unit.

The proton beam ranges in water calculated using Equation (1) are systematically lower than the d_max, d_d90, and d_d80 range values measured using the water tank setup. Applying a simple range calibration factor brings the MLFC range values to within ±0.4 g/cm² of d_max, as seen in Table 1. Figure 3A shows the MLFC derived and calibrated range as well as the d_max, d_d90, and d_d80 water tank range values.

Figure 3.

(A) Plot of multilayer Faraday collector (MLFC) derived and calibrated range as well as d_max, d_d90, and d_d80 water tank range values for 10 proton energies. Error bars show 95% confidence interval. (B) Plot of fortnightly measurements of MLFC peak channels for 10 energies across 11 weeks.

Table 1.

This Table Shows the Values for Multilayer Faraday Collector (MLFC) Calibrated Range and the Measured d_max, d_d90, and d_d80 Water Tank Range

Proton energy (MeV)	Calibrated MLFC range (cm)	Measured water tank range (cm)
Proton energy (MeV)	Calibrated MLFC range (cm)	d _max	d_d90	d_d80
70.2	3.6 ± 0.4	3.992	4.025	4.04
81.8	4.9 ± 0.4	5.319	5.361	5.382
94.6	6.6 ± 0.4	6.92	6.976	6.999
116.4	9.7 ± 0.5	10.038	10.111	10.142
142.4	14.2 ± 0.5	14.373	14.49	14.534
150.2	15.7 ± 0.5	15.762	15.89	15.943
168.0	19.1 ± 0.6	19.153	19.325	19.386
187.5	23.3 ± 0.6	23.202	23.39	23.461
208.3	28.1 ± 0.5	27.898	28.084	28.167
228.7	33.1 ± 0.8	32.778	33.003	33.099

For the purpose of QA, the reproducibility of the measured peak channel is as important as the accuracy of the derived range. Figure 3B shows the fortnightly variability of the measured MLFC peak channel. It was observed that the measured peak channel is consistent across 11 weeks for each energy, with the mean fortnightly measurements lying within the error bars of one another. The only outlier is the 150.2 MeV proton beam in week 7, which has been attributed to a beam tuning fault that week.

It was observed that the number of MUs and the SSD do not have a significant impact on the derived MLFC range, and that the range of the proton beam derived from the MLFC lie within 1 mm when these quantities were varied.

Charge

Figure 4A shows the comparison between the mean number of protons per MU calculated from the total charge collected by the MLFC fortnightly over 11 weeks against the value on our TPS. It was observed that the MLFC proton count agrees with the calculated proton count from the TPS within statistical uncertainties for low- and mid-range energies. For high-range energies, the MLFC measurements over-estimate the number of protons.

Figure 4.

(A) Plot of proton count per MU as measured by the MLFC with the theoretical value from the TPS. Error bars illustrate 1 standard deviation calculated from all measurements obtained across the weeks. (B) Plot of number of protons per MU measured by MLFC across 11 weeks, compared to the number of protons per MU given by TPS model. Abbreviations: TPS, treatment planning system; MLFC, multilayer Faraday collector; MU, monitor unit

We hypothesized 2 possible reasons for the discrepancy. The first possible reason is the effect of delayed charge,²² which happens when a small amount of charge is delivered from the PBT system after each spot. This may result in a difference between the planned and delivered MU per spot. We tested the integrity of MU per spot using log files obtained from the spot profile monitor (SPM) in the nozzle. From Figure 5A, it was observed that the delivered MU per spot deviates from the target of 0.2. This effect is more pronounced for the high-range energies. However, at the highest energy of 228.7 MeV, the maximum deviation is only 0.5% from the target, which is unlikely to account for the large discrepancy seen in Figure 4A.

Figure 5.

(A) Effects of delayed charge on the number of monitor unit (MU) per spot across the range of treatment energies. Error bars illustrate 1 standard deviation. (B) Effects of spill change on spot dose rate for 70.2, 150.2 and 228.7 MeV proton beams.

The second possible reason is the rapid drop in dose rate towards the end of the spill, which has been observed in our PBT system.²³ Hsi et al²⁴ has also reported a dose per MU (cGy/MU) dependence on d_d80 range, which is closely related to the energy of the protons. Through the SPM log files, it was observed that the spot dose rate is not constant and falls from the fixed target of 8 MU/s towards the end of the spills, as seen in Figure 5B. As such, Equation (3)'s assumption of a constant dose rate within the MLFC's 50-millisecond measurement interval time may not hold true.

To further study the discrepancy in total charge collected, we simulated the MLFC measurements using the SPM log files obtained from the same proton beams. As the SPM log files contain information on the spot dose rate (in MU/s) and the spot duration for each spot, the number of MU delivered by each spot can be determined. The cumulative number of MU delivered by all spots agree with the preset of 100 MU to within 0.3%, as shown in Table 2.

Table 2.

This Table Shows the Cumulative MU Simulated Using the SPM log File Before and After Resampling for 100 MU of 70.2, 150.2, and 228.7 MeV Proton Beams

Proton energy (MeV)	Cumulative MU from log file simulation	Cumulative MU after resampling
70.2	99.71	99.95
150.2	99.73	99.94
228.7	100.04	100.02

Abbreviations: MU, monitor unit; SPM, spot profile monitor.

We then investigated the effect of sampling time and varying dose rates on the cumulative MU delivered. As the SPM log files record actual delivered spot information for every spot, the log files were first resampled to a time interval equivalent to the MLFC integration time (ie $T_{mea} =$ 1 millisecond) and the spot dose rate were interpolated accordingly. Thereafter, the log files were downsampled to an interval equivalent to the MLFC measurement interval time (ie $T_{mea} + T_{dead} =$ 50 milliseconds) to simulate the MLFC measurement process. The cumulative number of MU delivered by all spots according to the resampled log file agree with the preset of 100 MU to within 0.1% as seen in Table 2, which cannot explain the large discrepancy seen in Figure 4A as well. Figure 6 shows a plot of the cumulative MU delivered by the proton beams across time, with the discontinuities representing the spill change.

Figure 6.

Plot of cumulative monitor unit (MU) delivered by 70.2, 150.2, and 228.7 MeV proton beams across time.

Finally, in Figure 4B, the time constancy of total charge collected by the MLFC was evaluated using the measurements taken fortnightly across 11 weeks.

Discussion

In this study, we demonstrated the use of an MLFC for proton beam range verification. The MLFC peak channel measured for the range of treatment energies scales linearly with the corresponding range measured with a water tank, indicating that the MLFC can be reliably used to determine the range of proton beams in water. While it is not as critical to obtain the absolute range values for the purpose of QA, applying a simple range calibration factor to the derived range provides a calibrated range that lies within 0.4 g/cm² of the d_max measured using a water tank setup.

The observed difference may be attributed, in part, to range straggling effects. Range straggling refers to the spread of proton energies within a beam, leading to a broadening of the Bragg peak.²⁵ As seen from Table 1 and Figure 3A, range straggling becomes more pronounced at higher proton energies, resulting in a broader energy deposition profile. However, it seems the that range straggling magnitude is smaller than the offset of 0.4 g/cm², so we can safely say that range straggling cannot fully account for the discrepancy observed in the measured range values.

Previous works that used MLFC for beam range verification, such as Tesfamicael et al,¹⁰ similarly applied range calibration factors to obtain range values comparable to water tank measurements. There could be several factors contributing to discrepancies between the MLFC's measurements and the actual range of the proton beam, such as errors in material stopping power (which was obtained from PSTAR in this work, instead of the MLFC manufacturer or derived experimentally), differences in energy losses through various processes such as ionization and scattering between the MLFC and the water tank, as well as uncertainties in layer thickness and composition of the MLFC. For these reasons, a range calibration factor may be necessary.

In Tesfamicael et al (2018)'s work, the water equivalent depth in the MLFC agreed to water tank d_d80 measurements to within 0.2 g/cm² after range calibration factors were applied. In this work, a simple range calibration factor that is energy-independent was used due to its simplicity and ease of application, which gives a calibrated range that agrees to 0.4 g/cm² of the water tank d_max measurements. In Testa et al's work, 2% agreement was found between validated TOPAS simulations and MLFC measurements, although no direct comparison was made between MLFC and water tank measurements.

Given the reproducibility of the range measurements across 11 weeks as well as its low dependence on set-up uncertainties (such as SSD) and number of MU used, the MLFC can be a fast and fuss-free method to verify the range of proton beams daily.

On charge verification, Figure 4A shows that the MLFC proton count agrees with the calculated proton count from the TPS within statistical uncertainties for low- and mid-range energies. While the effects of delayed charge and varying dose rates could have affected the charge collection at the high-range energies, our simulations of the MLFC measurements using the resampled SPM log files did not replicate the same discrepancies observed in our measurement data. This is likely because the time scale of the dose rate fluctuations due to the spill change is smaller than that of the MLFC measurements, hence it reproduces the same dose profile as the unsampled SPM log files.

One limitation of the MLFC's data acquisition system is the long deadtime between measurements. As the original intended use of the MLFC is for range or proton energy verification, the sampling time is unlikely to be an important parameter of consideration. Since the measurement deadtime has a significant impact on the charge collection,²⁶ an improved detector system with lower deadtime may yield more accurate results for charge verification. However, we do not foresee the sampling time to improve significantly in the near future for this product as the measurement of integrated charge of the entire proton irradiation is not part of the intended application. Finally, as there is variability in the charge collected over time, it suggests that the MLFC may only be suitable to catch gross errors in charge collection.

The findings in this work highlight the novelty and benefit of using the MLFC as a daily QA tool for range verification. It took less than 5 min on average to set up the MLFC, making it easier and faster to operate compared to a water tank and ionization chamber setup. While the MLFC may be unable to provide the absolute charge collected, it can be used to verify the collected charge following major maintenance changes to the beamline. This work is therefore useful for clinics looking for an alternative fuss-free method to perform QA checks on their PBT beamline.

Conclusion

This work demonstrates the application of an MLFC for proton beam range and charge collection verification for the purpose of PBT QA. For range measurements, the results demonstrate a strong correlation between the MLFC peak channel and the range values obtained from a water tank setup, indicating the MLFC's reliability in determining proton beam range. A simple energy-independent range calibration factor was applied, resulting in a calibrated range within 0.4 g/cm² of the d_max values measured with a water tank. The MLFC also demonstrates linearity, reproducibility, and negligible dependence on the number of MU and SSD used, making it a promising tool for daily range verification in proton therapy.

Regarding charge measurement, the MLFC proton count aligns well with values from the TPS for low- and mid-range energies, while overestimating at high-range energies. Possible reasons for this discrepancy include the effects of delayed charge and varying dose rates, although simulations did not fully replicate the observed discrepancies. The study suggests that the MLFC, while not providing absolute charge values, can be used to catch gross errors in charge collection, particularly following significant beamline maintenance changes. The findings highlight the MLFC's potential as a daily QA tool, offering a quicker and easier alternative to traditional water tank and ionization chamber setups. Further improvements in detector systems with lower deadtime could enhance the accuracy of charge measurements. This work contributes to the understanding of MLFC's role in proton therapy QA and offers insights for clinics seeking efficient QA methods for their proton beamlines.

Footnotes

Abbreviations

Authors’ Contributions

PLY, KSL, WYCK, and HQT contributed to the data acquisition and analysis. Statistical analyses were done by PLY and HQT. Administrative, technical, or material support was provided by PLY. Study supervision was done by JCLL and SYP. Drafting of the manuscript was done by PLY. Study conception and design, data interpretation, and approval of the final manuscript were done by all authors.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Duke-NUS Oncology Academic Program, Clinical & Systems Innovation Support—Innovation Seed Grant (grant number Goh Foundation Proton Research Programme: 08/FY202, 08/FY2022/P2/02-A68).

ORCID iDs

Ping L Yeap

Wei Y C Koh

References

Newhauser

Zhang

. The physics of proton therapy. Phys in Med Biol. 2015;60(8):R155.

Paganetti

(ed). Proton therapy physics. CRC press, 2018.

Yan

Ngoma

Ngwa

Bortfeld

. Global democratisation of proton radiotherapy. Lancet Oncol. 2023;24(6):e245‐e254.

Arjomandy

Taylor

Ainsley

, et al. AAPM Task group 224: Comprehensive proton therapy machine quality assurance. Med Phys. 2019;46(8):e678‐e705.

Newhauser

. International commission on radiation units and measurements report 78: Prescribing, recording and reporting proton-beam therapy. Radiat Prot Dosim. 2009;133(1):60‐62.

Arjomandy

Sahoo

Ding

, et al. Use of a two-dimensional ionization chamber array for proton therapy beam quality assurance. Med Phys. 2008;35(9):3889‐3894.

Vai

Mirandola

Magro

, et al. Characterization of a MLIC detector for QA in scanned proton and carbon ion beams. Int J of Particle Therapy. 2019;6(2):50‐59.

Ebara

Tsutsui

Nakajima

, et al. First beam extraction from a superconducting azimuthally varying field cyclotron for proton therapy. Nucl Instrum Methods Phys Res Sect A. 2023;1056(168629):1-10.

Gottschalk

Platais

Paganetti

. Nuclear interactions of 160 MeV protons stopping in copper: A test of monte carlo nuclear models. Med Phys. 1999;26(12):2597‐2601.

10.

Tesfamicael

Athar

Bejarano Buele

, et al. Use of commercial multilayer Faraday cup for offline daily beam range verification at the McLaren proton therapy center. Med Phys. 2019;46(2):1049‐1053.

11.

Boisseau

Dart

Gordon

, et al. 2016. Use of the Multi-layer Faraday Collector as a Convenient Transferable Standard for Proton Range and Beam Energy.

12.

Jung

Yoon

Moon

. Preliminary Test of Multi-layer Faraday Cup for Quality Assurance of Proton Beam Energy. Trans of the Korean Nucl Soc Virtual Spring Meeting. 2021. https://www.kns.org/files/pre_paper/45/21S-026-%EC%A0%95%EA%B4%91%EC%9D%BC.pdf

13.

Testa

Schümann

H-M

, et al. Experimental validation of the TOPAS Monte Carlo system for passive scattering proton therapy. Med Phys. 2013;40(12):121719.

14.

Pyramid Technical Consultants, Inc. MLFC-128-250 Datasheet. Available at: https://ptcusa.com/products/mlfc-128 (Accessed: 19 April 2024)

15.

Tsubouchi

Beltran

Yagi

, et al. Beam delivery characteristics of the hitachi carbon ion scanning system at Osaka heavy Ion medical accelerator in kansai (HIMAK). Med Phys. 2023;51(3):2239‐2250.

16.

Cruz

Narayanasamy

Papanikolaou

Stathakis

. Dosimetric comparison of water phantoms, ion chambers, and data acquisition modes for LINAC characterization. Radiat Meas. 2015;82:108‐114.

17.

Kuess

Haupt

Osorio

, et al. Characterization of the PTW-34089 type 147 mm diameter large-area ionization chamber for use in light-ion beams. Phys in Med Biol. 2020;65(17):17NT02.

18.

Zhang

Newhauser

. Calculation of water equivalent thickness of materials of arbitrary density, elemental composition and thickness in proton beam irradiation. Phys in Med Biol. 2009;54(6):1383‐1395.

19.

Berger

Coursey

Zucker

Chang

. ESTAR, PSTAR, and ASTAR: Computer Programs for Calculating Stopping-Power and Range Tables for Electrons, Protons, and Helium Ions (version 1.2.3). [Online] Available: http://physics.nist.gov/Star [2023, August 27]. National Institute of Standards and Technology, Gaithersburg, MD, 2005.

20.

Palmans

Vatnitsky

. Beam monitor calibration in scanned light-ion beams. Med Phys. 2016;43(11):5835‐5847.

21.

Osorio

Dreindl

Grevillot

, et al. Beam monitor calibration of a synchrotron-based scanned light-ion beam delivery system. Zeitschrift für Medizinische Physik. 2021;31(2):154‐165.

22.

Whitaker

Beltran

Tryggestad

, et al. Comparison of two methods for minimizing the effect of delayed charge on the dose delivered with a synchrotron based discrete spot scanning proton beam. Med Phys. 2014;41(8):081703.

23.

Tan

Lew

Koh

CWY

, et al. The effect of spill change on reliable absolute dosimetry in a synchrotron proton spot scanning system. Med Phys. 2023;50(7):4067‐4078.

24.

Hsi

Schreuder

Moyers

, et al. Range and modulation dependencies for proton beam dose per monitor unit calculations. Med Phys. 2009;36(2):634‐641.

25.

Russell

Grusell

Montelius

. Dose calculations in proton beams: Range straggling corrections and energy scaling. Phys Med Biol. 1995;40(6):1031‐1043.

26.

Patera

Sarti

. Recent advances in detector technologies for particle therapy beam monitoring and dosimetry. IEEE Trans on Radiat Plasma Med Sci. 2020;4(2):133‐146.