Fig 1 - uploaded by Chan Hyeong Kim
Content may be subject to copyright.
Source publication
To verify in-vivo proton dose distribution, a 2-dimensional (2D) prompt-gamma measurement system, comprised of a multi-hole collimation system, a 2D array of CsI(Tl) scintillators, and a position-sensitive photomultiplier tube (PS-PMT), is under development. In the present study, to determine the optimal dimension of the measurement system, we empl...
Contexts in source publication
Context 1
... is impossible to directly measure them for verification of the proton dose or the beam range; hence, measuring the prompt gammas generated by proton-induced nu- clear interactions has been suggested [9,10]. Recently, a close correlation between the longitudinal distribution of prompt gammas and the location of the distal dose fall-off was shown by using a prompt gamma scanning (PGS) system that provided one-dimensional infor- mation on the prompt-gamma distribution [11]. As an extension of that study, a 2-dimensional (2D) prompt- gamma measurement system consisting of a multi-hole tungsten collimator, a 2D array of CsI(Tl) scintillators, and a commercial position-sensitive photomultiplier tube (PS-PMT) currently is under development. The hope is that the developed system will provide the 2D prompt- gamma distribution, which could then be correlated with the proton dose distribution in the patient. In the present study, the optimal dimensions of the 2D prompt-gamma measurement system were determined by using Monte Carlo simulations with the MCNPX code [12]. To enable measurement of prompt-gamma distribution while effectively discriminating the background gammas generated mostly by neutron capture processes, we determined the optimal dimensions of the multi-hole collimator and the scintillators. Finally, we determined the performance of the measurement system, as configured for the optimal dimensions, was estimated for 80-, 150-, and 200-MeV proton beams incident on a 20 cm × 20 cm × 40 cm water phantom in the longitudinal direction. If the 2D distribution of prompt gammas is to be mea- sured, it is important to effectively shield background gammas and measure only prompt gammas from the proton beam‘s passage. Furthermore, for measurement of high-energy prompt gammas, the CsI(Tl) scintillator must not be too small; otherwise, high-energy secondary electrons generated by the prompt gammas can escape, resulting in serious deterioration of the detector function. Thus, the dimensions of the multi-hole collimation system and the CsI(Tl) scintillator are important design parameters subject to optimization. In the present study, the collimator hole size, collimator thickness, and scintillator length (see Fig. 1) were optimized by using Monte Carlo simulations with the MCNPX code. The LA150 data libraries, which describe evaluated proton, neutron, and photonuclear cross-section data up to 150 MeV, and the ISABEL model were used to model the proton interactions in a water phantom. With the LA150 data libraries, the MCNPX code can precisely simulate proton-induced nu- clear interactions and secondary-particle interactions up to 150 MeV [13,14]. For protons above 150 MeV, al- though the MCNPX code uses both the data library and physics models for particle transportation, it does not significantly affect the results considering that the proton energy is below 150 MeV near the Bragg-peak region and that there is little difference in background gammas generated by neutron captures. The cross-sectional area of the CsI(Tl) scintillator, one of the most important design parameters of the measure- ment system, was varied identically with the collimator hole size (see Fig. 1); that is, it was not optimized sep- arately. The pitch of the collimator holes was fixed at 6 mm under the assumption that the developed system will use a commercial PS-PMT (H8500C, Hamamatsu Photonics K.K., Japan), the pitch of which is 6 mm. For optimization, 1.21 × 10 9 150-MeV protons, correspond- ing to 10 spots in spot scanning treatment [15], were delivered as a pencil beam to the center of the 20 cm × 20 cm × 40 cm water phantom in the longitudinal direction, after which the prompt-gamma distribution was simulated varying the collimator hole size, collimator thickness, and scintillator length. To effectively discrimi- nate the background gammas from the prompt gammas, we applied an optimal energy window of 4 – 10 MeV [16] to the scintillation detectors. The optimal dimensions of the measurement system were then determined by monitoring the prompt-gamma distributions in the lateral and longitudinal directions. In the case of the longitudinal direction, the peak-to-background (PB) ratio, defined as the ratio of the gamma counts at the peak to the average background gamma counts within a 2- to 4-cm range distal to the peak, was used. Based on results of the optimization studies, a 2D prompt-gamma measurement system was designed, and its performance was predicted using the MCNPX code. This involved delivery of monoenergetic proton beams of 80, 150, and 200 MeV to the center of the water phantom in the longitudinal direction. The 2D distributions of prompt gammas were simulated for the optimized measurement system. First, the prompt-gamma distribution was simulated for various hole sizes of the multi-hole collimation system. Figure 2(a) shows the lateral distribution of the simulated prompt gammas as a function of the collimator hole size. The result indicates that a 2D prompt- gamma measurement system cannot accurately measure the prompt-gamma distribution if the collimator hole size is too small ( i.e. , less than 0.3 × 0.3 cm 2 ). In this case, not only are the majority of the prompt gammas shielded by the collimator but also the energies of the prompt gammas are incompletely absorbed in the CsI(Tl) scintillators due to the escape of high-energy secondary electrons. Figure 2(b) plots the PB ratio as a function of the collimator hole size. Whereas the ratio can be seen to increases gradually with the collimator hole size, it falls rapidly once the hole size exceeds 0.4 × 0.4 cm 2 . Such hole sizes excessively reduce the sep- tum thickness for the fixed pitch (= 0.6 cm), rendering the collimation system ineffective for shielding against background gammas and prompt gammas in unwanted directions. Next, the prompt-gamma distribution was simulated for various collimator thicknesses. Figure 3(a) shows the lateral distribution of prompt gammas at the peak location as a function of the collimator thickness. The result clearly indicates that the relative contribution of the background gammas decreases with increasing collimator thickness, but that the variation becomes almost negligible if the thickness is equal to or greater than 15 cm. Figure 3(b) plots the variation of the PB ratio as a function of the collimator thickness. Whereas the ratio increases with increasing collimator thickness for thicknesses less than 15 cm, it drops sharply for thicknesses greater than 15 cm due to the fact that at collimator thicknesses greater than 15 cm, the reduction in the prompt gammas has more influence on the simulated- gamma distribution than the reduction in background gammas dose. Figure 4(a) shows the number of gamma counts at the peak location as a function of the scintillator length. The peak counts dramatically increase until the scintillator length reaches 5 cm, after which, however, they increase only rather slowly. Figure 4(b) indicates, similarly, that the PB ratio increases with the scintillator length until the scintillator length reaches 5 cm, and then decreases. This declination of the PB ratio is understand- able, considering that after 5 cm, the increase in background gamma counts due to the increase in the scintillator length is even greater than that of the prompt-gamma counts. Based on a summary of the simulation results, the optimal dimensions of the 2D prompt-gamma measurement system were determined to be 0.4 × 0.4 cm 2 , 15 cm, and 5 cm for the collimator hole size, the collimator thickness, and the scintillator length, respectively. Finally, the performance of the 2D prompt-gamma measurement system, as configured for the optimal dimensions so determined, was estimated for 80-, 150-, and 200-MeV proton beams. The distance between the water phantom and the measurement system was 10 cm, and the number of protons was, again, 1.21 × 10 9 for each simulation. Figure 5 plots the 2D distributions of the simulated prompt gammas for the optimized measurement system and the depth distributions of prompt gammas at the incident positions of the proton beams. The results show that the measurement system can identify the peak location of prompt gammas near the end of the proton beam range for the 80- and the 150-MeV proton beams, but not for the 200-MeV proton beam. In the 200-MeV case, the failure to determine the peak location was attributed to the dominance of the background gammas and to the lateral dispersion of the proton beam at the end of the beam range. The present study employed a series of Monte Carlo simulations with the MCNPX code to determine the optimal design of a 2D prompt-gamma measurement system comprised of a multi-hole collimation system, a 2D array of CsI(Tl) scintillators, and a PS-PMT. The optimal dimensions of the system were determined to be 0.4 × 0.4 cm 2 , 15 cm, and 5 cm for the collimator hole size, collimator thickness, and scintillator length, respectively. Thereafter, the system, so configured, was simulated, the results showing accurate measurement of the 2D prompt-gamma distribution along with effective de- terminations of the peak locations of prompt gammas near the end of the proton beam range for 80- and 150-MeV proton beams, but not for a 200-MeV proton beam. Based on these simulation results, a prototype 2D prompt-gamma measurement system currently is under construction and, upon completion, will be tested with therapeutic proton beams. This research was supported by the National Nuclear R&D Program through the National Research Founda- tion of Korea (NRF) funded by the Ministry of Educa- tion, Science and Technology (Nos. 2010-0028913, 2010- 0023825, and ...
Context 2
... is impossible to directly measure them for verification of the proton dose or the beam range; hence, measuring the prompt gammas generated by proton-induced nu- clear interactions has been suggested [9,10]. Recently, a close correlation between the longitudinal distribution of prompt gammas and the location of the distal dose fall-off was shown by using a prompt gamma scanning (PGS) system that provided one-dimensional infor- mation on the prompt-gamma distribution [11]. As an extension of that study, a 2-dimensional (2D) prompt- gamma measurement system consisting of a multi-hole tungsten collimator, a 2D array of CsI(Tl) scintillators, and a commercial position-sensitive photomultiplier tube (PS-PMT) currently is under development. The hope is that the developed system will provide the 2D prompt- gamma distribution, which could then be correlated with the proton dose distribution in the patient. In the present study, the optimal dimensions of the 2D prompt-gamma measurement system were determined by using Monte Carlo simulations with the MCNPX code [12]. To enable measurement of prompt-gamma distribution while effectively discriminating the background gammas generated mostly by neutron capture processes, we determined the optimal dimensions of the multi-hole collimator and the scintillators. Finally, we determined the performance of the measurement system, as configured for the optimal dimensions, was estimated for 80-, 150-, and 200-MeV proton beams incident on a 20 cm × 20 cm × 40 cm water phantom in the longitudinal direction. If the 2D distribution of prompt gammas is to be mea- sured, it is important to effectively shield background gammas and measure only prompt gammas from the proton beam‘s passage. Furthermore, for measurement of high-energy prompt gammas, the CsI(Tl) scintillator must not be too small; otherwise, high-energy secondary electrons generated by the prompt gammas can escape, resulting in serious deterioration of the detector function. Thus, the dimensions of the multi-hole collimation system and the CsI(Tl) scintillator are important design parameters subject to optimization. In the present study, the collimator hole size, collimator thickness, and scintillator length (see Fig. 1) were optimized by using Monte Carlo simulations with the MCNPX code. The LA150 data libraries, which describe evaluated proton, neutron, and photonuclear cross-section data up to 150 MeV, and the ISABEL model were used to model the proton interactions in a water phantom. With the LA150 data libraries, the MCNPX code can precisely simulate proton-induced nu- clear interactions and secondary-particle interactions up to 150 MeV [13,14]. For protons above 150 MeV, al- though the MCNPX code uses both the data library and physics models for particle transportation, it does not significantly affect the results considering that the proton energy is below 150 MeV near the Bragg-peak region and that there is little difference in background gammas generated by neutron captures. The cross-sectional area of the CsI(Tl) scintillator, one of the most important design parameters of the measure- ment system, was varied identically with the collimator hole size (see Fig. 1); that is, it was not optimized sep- arately. The pitch of the collimator holes was fixed at 6 mm under the assumption that the developed system will use a commercial PS-PMT (H8500C, Hamamatsu Photonics K.K., Japan), the pitch of which is 6 mm. For optimization, 1.21 × 10 9 150-MeV protons, correspond- ing to 10 spots in spot scanning treatment [15], were delivered as a pencil beam to the center of the 20 cm × 20 cm × 40 cm water phantom in the longitudinal direction, after which the prompt-gamma distribution was simulated varying the collimator hole size, collimator thickness, and scintillator length. To effectively discrimi- nate the background gammas from the prompt gammas, we applied an optimal energy window of 4 – 10 MeV [16] to the scintillation detectors. The optimal dimensions of the measurement system were then determined by monitoring the prompt-gamma distributions in the lateral and longitudinal directions. In the case of the longitudinal direction, the peak-to-background (PB) ratio, defined as the ratio of the gamma counts at the peak to the average background gamma counts within a 2- to 4-cm range distal to the peak, was used. Based on results of the optimization studies, a 2D prompt-gamma measurement system was designed, and its performance was predicted using the MCNPX code. This involved delivery of monoenergetic proton beams of 80, 150, and 200 MeV to the center of the water phantom in the longitudinal direction. The 2D distributions of prompt gammas were simulated for the optimized measurement system. First, the prompt-gamma distribution was simulated for various hole sizes of the multi-hole collimation system. Figure 2(a) shows the lateral distribution of the simulated prompt gammas as a function of the collimator hole size. The result indicates that a 2D prompt- gamma measurement system cannot accurately measure the prompt-gamma distribution if the collimator hole size is too small ( i.e. , less than 0.3 × 0.3 cm 2 ). In this case, not only are the majority of the prompt gammas shielded by the collimator but also the energies of the prompt gammas are incompletely absorbed in the CsI(Tl) scintillators due to the escape of high-energy secondary electrons. Figure 2(b) plots the PB ratio as a function of the collimator hole size. Whereas the ratio can be seen to increases gradually with the collimator hole size, it falls rapidly once the hole size exceeds 0.4 × 0.4 cm 2 . Such hole sizes excessively reduce the sep- tum thickness for the fixed pitch (= 0.6 cm), rendering the collimation system ineffective for shielding against background gammas and prompt gammas in unwanted directions. Next, the prompt-gamma distribution was simulated for various collimator thicknesses. Figure 3(a) shows the lateral distribution of prompt gammas at the peak location as a function of the collimator thickness. The result clearly indicates that the relative contribution of the background gammas decreases with increasing collimator thickness, but that the variation becomes almost negligible if the thickness is equal to or greater than 15 cm. Figure 3(b) plots the variation of the PB ratio as a function of the collimator thickness. Whereas the ratio increases with increasing collimator thickness for thicknesses less than 15 cm, it drops sharply for thicknesses greater than 15 cm due to the fact that at collimator thicknesses greater than 15 cm, the reduction in the prompt gammas has more influence on the simulated- gamma distribution than the reduction in background gammas dose. Figure 4(a) shows the number of gamma counts at the peak location as a function of the scintillator length. The peak counts dramatically increase until the scintillator length reaches 5 cm, after which, however, they increase only rather slowly. Figure 4(b) indicates, similarly, that the PB ratio increases with the scintillator length until the scintillator length reaches 5 cm, and then decreases. This declination of the PB ratio is understand- able, considering that after 5 cm, the increase in background gamma counts due to the increase in the scintillator length is even greater than that of the prompt-gamma counts. Based on a summary of the simulation results, the optimal dimensions of the 2D prompt-gamma measurement system were determined to be 0.4 × 0.4 cm 2 , 15 cm, and 5 cm for the collimator hole size, the collimator thickness, and the scintillator length, respectively. Finally, the performance of the 2D prompt-gamma measurement system, as configured for the optimal dimensions so determined, was estimated for 80-, 150-, and 200-MeV proton beams. The distance between the water phantom and the measurement system was 10 cm, and the number of protons was, again, 1.21 × 10 9 for each simulation. Figure 5 plots the 2D distributions of the simulated prompt gammas for the optimized measurement system and the depth distributions of prompt gammas at the incident positions of the proton beams. The results show that the measurement system can identify the peak location of prompt gammas near the end of the proton beam range for the 80- and the 150-MeV proton beams, but not for the 200-MeV proton beam. In the 200-MeV case, the failure to determine the peak location was attributed to the dominance of the background gammas and to the lateral dispersion of the proton beam at the end of the beam range. The present study employed a series of Monte Carlo simulations with the MCNPX code to determine the optimal design of a 2D prompt-gamma measurement system comprised of a multi-hole collimation system, a 2D array of CsI(Tl) scintillators, and a PS-PMT. The optimal dimensions of the system were determined to be 0.4 × 0.4 cm 2 , 15 cm, and 5 cm for the collimator hole size, collimator thickness, and scintillator length, respectively. Thereafter, the system, so configured, was simulated, the results showing accurate measurement of the 2D prompt-gamma distribution along with effective de- terminations of the peak locations of prompt gammas near the end of the proton beam range for 80- and 150-MeV proton beams, but not for a 200-MeV proton beam. Based on these simulation results, a prototype 2D prompt-gamma measurement system currently is under construction and, upon completion, will be tested with therapeutic proton beams. This research was supported by the National Nuclear R&D Program through the National Research Founda- tion of Korea (NRF) funded by the Ministry of Educa- tion, Science and Technology (Nos. 2010-0028913, 2010- 0023825, and ...
Similar publications
An accurate knowledge of in vivo proton dose distribution is key to fully utilizing the potential advantages of proton therapy. Two representative indirect methods for in vivo range verification, namely, prompt gamma (PG) imaging and positron emission tomography (PET), are available. This study proposes a PG-PET system that combines the advantages...
Online ion range monitoring in hadron therapy can be performed via detection of secondary radiation, such as prompt γ-rays, emitted during treatment. The prompt γ emission profile is correlated with the ion depth-dose profile and can be reconstructed via Compton imaging. The line-cone reconstruction, using the intersection between the primary beam...
In proton therapy, verification of proton dose distribution is important to treat cancer precisely and to enhance patients’ safety. To verify proton dose distribution, in a previous study, our team incorporated a vertically-aligned one-dimensional array detection system. We measured 2D prompt-gamma distribution moving the developed detection system...
Citations
... Testa et al. (2008) used the same method for carbon ion irradiation. In addition, a two dimensional parallel hole system was modeled in MCNPX Monte-Carlo code by Lee et al. (2012). A Parallel slit system was also modeled in MCNPX code by Min et al. (2012). ...
In Radiation Oncology, proton therapy has become an increasingly popular treatment modality due to the superior dose distribution of the proton beam while sparing more surrounding normal healthy tissues and critical organs. This advantage can quickly turn into a disadvantage if there is any uncertainty in the delivery of the proton beam. To fully utilize the benefits of proton therapy, it is important to monitor the in-vivo dose deposition. Due to the fact that the treatment protons stop within the patient as they deliver the dose, secondary radiation is the potential method to obtain a dose verification measurement. The detection of secondary prompt gamma rays have been proposed as an in-situ method to determine the proton range since the location of the prompt gamma emission is strongly correlated with the proton depth dose profile. This correlation has been confirmed in both experimental measurements and in Monte Carlo simulations, but absolute prompt gamma productions have been unsuccessful, due to discrepancies the Monte Carlo prompt gamma production data particularly for the prominent elements found in tissue within the therapeutic range (50-200 MeV).
The goal of this work was to evaluate the prompt gamma production for both carbon and oxygen at energies relevant for proton therapy. The first part of this study was to experimentally measure the interaction cross section for proton-nucleus collisions in both carbon and oxygen. In order to determine these cross-sections, measurements using thin targets of natural Carbon and Mylar over the energy range of 66-125 MeV were performed using the AFRODITE detector system at iThemba LABS in Cape Town, South Africa. Energy and efficiency calibrations of the detection system were performed using three standard gamma emitting sources (137Cs, 60Co, and 152Eu). The second part of this work was to model the AFRODITE detector system using the Geant4 Monte-Carlo radiation transport code in order to compare the simulated to the measured results and to evaluate the previously observed discrepancies for prompt gamma production in the Geant4 code.
In the experimental study, the standard gamma sources were used to obtain individual absolute efficiency response curves for each of the 30 high purity germanium crystals (2 crystals were not functioning). The differential cross section was calculated at five different energies (66, 80, 95, 110, 125 MeV) using three angles (90deg, 130deg and 140deg) for the 4.438 MeV gamma peak using both the carbon and Mylar targets, and for the 6.129 MeV gamma peak using the Mylar target. The total cross-section for the 4.438 and 6.129 MeV peaks were calculated by fitting the angular measurements to a Legendre polynomial.
In the simulation study, the geometry of the AFRODITE detection system was carefully modelled to mimic the actual geometry by importing CAD models into the Geant4 code. The physics of the AFRODITE model was tested by comparison to the three standard gamma emitting sources, by testing the Compton suppression system and evaluating var- ious hadronic physics processes. Once the model was validated, the experimental runs were simulated and the same procedures were followed in order to obtain absolute detector efficiency curves for the germanium crystals, as well as, differential cross-section data and total cross-section data for the 4.438 and 6.129 MeV gamma peaks.
The overall absolute gamma energy spectra from the experimental and simulation runs were compared and displayed excellent agreement for the total prompt gamma production values. Comparisons of the cross-section values for the individual peaks (4.438 and 6.129 MeV) produced mixed results. There was strong agreement (difference of 1.2% to 14.0%) for the total experimental cross-section results for the 4.438 MeV peak from the two targets (carbon and Mylar), providing internal validation for the experimental measurements. However, the total cross-section data for both peaks was slightly higher than the sparse previously available measured data points.
When comparing to the simulated results, the cross-section values for the individual peaks (4.438 and 6.129 MeV) did not agree quite as well. The simulated results for the total cross-section for the 4.438 MeV peak from the carbon target, the 12C(p, p′)12C* reaction, produced values 50% higher than the measured results. This over-estimation has been attributed to an unusually high number of 12C(p, d)11C* and 12C(p, np)11C* reactions produced during the simulation. The total cross-section for the 4.438 MeV peak from the Mylar target, the 12C(p, p′)12C* reaction, also produced an over-estimation (32%) for the same reason; this result was also boosted by the 16O(p, x)12C* reactions that occur from the oxygen found in Mylar. The 6.129 MeV peak from the Mylar target, the 16O(p, p′)16O* reaction, was virtually non-existent in the simulations making it impossible to calculate a simulated total cross-section value.
Overall, the experimental measured data provided results that compare favourably to ex- isting data. The available experimental data for the 12C(p, p′)12C* and 16O(p, p′)16O* reactions has been extended up to 125 MeV. The AFRODITE Geant4 model successfully re-produced the gamma spectra from the experimental runs and will be used again in further studies. As seen before, the Geant4 physics over-estimates (or neglects) the individual prompt gamma peaks. Further testing will be done to identify these errors and work to improve the Geant4 physics models.
... A pin-hole camera (30) is the pioneer approach to scan the prompt gamma emission distribution in a right angle to the beam track. Many research groups have performed experiments based on slit cameras at proton or carbon beams (31)(32)(33)(34)(35)(36)(37). The knife-edgeshaped camera has demonstrated the feasibility of millimeter range verification at clinical current intensities (38) in real time on a spot basis with realistic treatment plans and heterogeneous phantoms (39). ...
Proton beams are promising means for treating tumors. Such charged particles stop at a defined depth, where the ionization density is maximum. As the dose deposit beyond this distal edge is very low, proton therapy minimizes the damage to normal tissue compared to photon therapy. Nevertheless, inherent range uncertainties cast doubts on the irradiation of tumors close to organs at risk and lead to the application of conservative safety margins. This constrains significantly the potential benefits of protons over photons. In this context, several research groups are developing experimental tools for range verification based on the detection of prompt gammas, a nuclear by-product of the proton irradiation. At OncoRay and Helmholtz-Zentrum Dresden-Rossendorf, detector components have been characterized in realistic radiation environments as a step toward a clinical Compton camera. On the one hand, corresponding experimental methods and results obtained during the ENTERVISION training network are reviewed. On the other hand, a novel method based on timing spectroscopy has been proposed as an alternative to collimated imaging systems. The first tests of the timing method at a clinical proton accelerator are summarized, its applicability in a clinical environment for challenging the current safety margins is assessed, and the factors limiting its precision are discussed.
... Among the proposed designs, we can mention SPECT systems such as knife-edge-slit, pinhole and parallel-slit collimator cameras (e.g. Bom et al (2012), Kim et al (2009), Lee et al (2012), Cambraia Lopes et al (2012), Perali et al (2014), Smeets et al (2012), Verburg and Seco (2014)), as well as Compton cameras (e.g. Frandes et al (2010), Kormoll et al (2011), Kurosawa et al (2012) and Richard et al (2012)). ...
In proton therapy, the prompt-γ (PG) radiation produced by the interactions between protons and matter is related to the range of the beam in the patient. Tomographic Compton imaging is currently studied to establish a PG image and verify the treatment. However the quality of the reconstructed images depends on a number of factors such as the volume attenuation, the spatial and energy resolutions of the detectors, incomplete absorptions of high energy photons and noise from other particles reaching the camera. The impact of all these factors was not assessed in details. In this paper we investigate the influence of the PG energy spectrum on the reconstructed images. To this aim, we describe the process from the Monte Carlo simulation of the proton irradiation, through the Compton imaging of the PG distribution, up to the image reconstruction with a statistical MLEM method. We identify specific PG energy windows that are more relevant to detect discrepancies with the treatment plan. We find that for the simulated Compton device, the incomplete absorption of the photons with energy above about 2 MeV prevents the observation of the PG distributions at specific energies. It also leads to blurred images and smooths the distal slope of the 1D PG profiles obtained as projections on the central beam axis. We show that a selection of the events produced by γ photons having deposited almost all their energy in the camera allows to largely improve the images, a result that emphasizes the importance of the choice of the detector. However, this initial-energy-based selection is not accessible in practice. We then propose a method to estimate the range of the PG profile both for specific deposited-energy windows and for the full spectrum emission. The method relies on two parameters. We use a learning approach for their estimation and we show that it allows to detect few millimeter shifts of the PG profiles.
... Several designs and concepts resorting to prompt-gamma detection have been proposed for ion beam range monitoring, namely the multi-slit collimated camera (Min et al 2006, Testa et al 2008), the Compton camera (Peterson et al 2010, Kormoll et al 2011, Richard et al 2011, Robertson et al 2011, Park et al 2012), the knife-edge-shaped slit camera (Bom et al 2012, Smeets et al 2012), the pinhole camera (Kim et al 2009), the multi-hole camera (Lee et al 2012), the prompt-gamma timing (Golnik et al 2014), and the single-slit collimator with energy and time-resolved detection of prompt gammas (Verburg et al 2013). Specifically for collimated cameras, studies have shown that it is possible to retrieve information about the ion range in real time both for protons (Min et al 2006) and carbon ions (Testa et al 2008, Testa et al 2009). ...
Prompt-gamma emission detection is a promising technique for hadrontherapy monitoring purposes. In this regard, obtaining prompt-gamma yields that can be used to develop monitoring systems based on this principle is of utmost importance since any camera design must cope with the available signal. Herein, a comprehensive study of the data from ten single-slit experiments is presented, five consisting in the irradiation of either PMMA or water targets with lower and higher energy carbon ions, and another five experiments using PMMA targets and proton beams. Analysis techniques such as background subtraction methods, geometrical normalization, and systematic uncertainty estimation were applied to the data in order to obtain absolute prompt-gamma yields in units of prompt-gamma counts per incident ion, unit of field of view, and unit of solid angle. At the entrance of a PMMA target, where the contribution of secondary nuclear reactions is negligible, prompt-gamma counts per incident ion, per millimetre and per steradian equal to (124 ± 0.7stat ± 30sys) × 10(-6) for 95 MeV u(-1) carbon ions, (79 ± 2stat ± 23sys) × 10(-6) for 310 MeV u(-1) carbon ions, and (16 ± 0.07stat ± 1sys) × 10(-6) for 160 MeV protons were found for prompt gammas with energies higher than 1 MeV. This shows a factor 5 between the yields of two different ions species with the same range in water (160 MeV protons and 310 MeV u(-1) carbon ions). The target composition was also found to influence the prompt-gamma yield since, for 300/310 MeV u(-1) carbon ions, a 42% greater yield ((112 ± 1stat ± 22sys) × 10(-6) counts ion(-1) mm(-1) sr(-1)) was obtained with a water target compared to a PMMA one.
... The main challenges are the relatively high energies and count rates as well as the elevated neutron background. Several designs have been proposed to image prompt gamma, amongst them pinhole, knife-edge-slit and parallel-slit collimator designs (see Kim 2009, Testa et al 2010, Lee et al 2012, Smeets et al 2012, Diblen et al 2012, Bom et al 2012, Lopes et al 2012 as well as electronic collimation via Compton cameras (see Richard et al 2011, Peterson Testa et al (2008) to decrease the neutron-induced background, and enable the observation of prompt gamma for carbon ions. The parallel-slit collimator is a promising and straightforward solution for prompt-gamma imaging that does not have to contend with the complexity of, for example, a Compton imaging system. ...
Prompt-gamma profile was measured at WPE-Essen using 160 MeV protons impinging a movable PMMA target. A single collimated detector was used with time-of-flight (TOF) to reduce the background due to neutrons. The target entrance rise and the Bragg peak falloff retrieval precision was determined as a function of incident proton number by a fitting procedure using independent data sets. Assuming improved sensitivity of this camera design by using a greater number of detectors, retrieval precisions of 1 to 2 mm (rms) are expected for a clinical pencil beam. TOF improves the contrast-to-noise ratio and the performance of the method significantly.
... In vivo range verification is one of the methods utilized for range uncertainty reduction in patients. Several such studies have been carried out using positron emission tomography (PET) imaging, [6][7][8][9][10][11] prompt gamma timing (PGT), 12,13 prompt gamma peak integral (PGPI), 14 prompt gamma spectroscopy (PGS), 15 prompt gamma imaging (PGI), 4,[16][17][18][19][20][21][22][23][24][25][26][27][28][29][30] and others. PGI is one of the most promising techniques, because the prompt gamma (PG) distribution is strongly correlated with the proton dose distribution and PGs are generated within a few nanoseconds, which enables real-time verification during treatment. ...
Background
In proton therapy, a highly steep distal dose penumbra can be utilized for dose conformity, given the Bragg peak characteristic of protons. However, the location of the Bragg peak in patients (i.e., the beam range) is very sensitive to range uncertainty. Even a small shift of beam range can produce a significant variation of delivered dose to tumor and normal tissues, thus degrading treatment quality and threatening patient safety. This range uncertainty issue, therefore, is one of the important aspects to be managed in proton therapy.
Purpose
For better management of range uncertainty, range verification has been widely studied, and prompt gamma imaging (PGI) is considered one of the promising methods in that effort. In this context, a PGI system named the gamma electron vertex imaging (GEVI) system was developed and recently upgraded for application to pencil‐beam scanning (PBS) proton therapy. Here, we report the first experimental results using the therapeutic spot scanning proton beams.
Methods
A homogeneous slab phantom and an anthropomorphic phantom were employed. Spherical and cubic planning target volumes (PTVs) were defined. Various range shift scenarios were introduced. Prompt gamma (PG) measurement was synchronized with beam irradiation. The measured PG distributions were aggregated to improve the PG statistics. The range shift was estimated based on the relative change of the centroid in the measured PG distribution. The estimated range shifts were analyzed by range shift mapping, confidence interval (CI) estimation, and statistical hypothesis testing.
Results
The range shift mapping results showed an obvious measured range shift tendency following the true shift values. However, some fluctuations were found for spots that had still‐low PG statistics after spot aggregation. The 99% CI distributions showed clearly distributed range shift measurement data. The overall accuracy and precision for all investigated scenarios were 0.36 and 0.20 mm, respectively. The results of one‐sample t‐tests confirmed that every shift scenario could be observed up to 1 mm of shift. The ANOVA results proved that the measured range shift data could be discriminated from one another, except for 16 (of 138) comparison cases having 1–2 mm shift differences.
Conclusions
This study demonstrated the feasibility of the GEVI system for measurement of range shift in spot scanning proton therapy. Our experimental results showed that the proton beam can be measured up to 1 mm of range shift with high accuracy and precision. We believe that the GEVI system is one of the most promising PGI systems for in vivo range verification. Further research for application to more various cases and patient treatments is planned.
Multi-slat prompt-gamma camera is a promising tool for range monitoring during proton therapy. We report the results of a comprehensive simulation study analyzing the precision which is possible to reach with this camera in determination of the position of the distal edge of the Bragg peak. For the first time we include simulation of optical photons. The proton beam (single pencil beam, 130 MeV, 10 ns bunch period, total of 1·10⁸ protons) is interacting with a polymethyl methacrylate (PMMA) phantom, which is a cylinder of 200 mm in diameter and length. The prompt gamma rays generated in the phantom are collimated with a multi-slat collimator and detected using a combination of yttrium aluminum perovskite (YAP) scintillators, installed in the collimator apertures, and light sensors. Two scintillator packing schemes, with one and with two scintillator plates per aperture, are considered. The collimator configuration (the septal thickness, aperture and height), resulting in the best precision, is determined using two methods of detector optimization. Precision of 2.1 mm (full width at half maximum) in the edge position determination is demonstrated.
In the present study, a new prompt-gamma activation imaging system was proposed for two-dimensional (2D) elemental distributions. The system, as based on coincidence measurement, consists of a high-purity germanium detector for the measurement of the prompt-gamma energy and a position-sensitive detector for the determination of the emission position. To estimate its feasibility, we performed the Monte Carlo simulations using the Geant4 toolkit. For an iron bulk sample with implanted nickel elements, we could recognize the shape of the nickel implant from the 2D image obtained by the imaging system using the energy information of characteristic prompt gammas emitted by neutron capture interactions.
In proton therapy, accurate monitoring of the in-vivo proton dose distribution is essential in order to deliver the planned dose to the tumor volume within a minimal safety margin. Recently, a strong correlation between the distributions of the proton dose and the prompt gammas was found, and various prompt-gamma distribution-measurement systems, including collimation-based systems, Compton cameras, knife-edge imaging systems, and ion vertex imaging systems, have been proposed. In the present study, the feasibility of proton dose distribution monitoring was tested using a two-dimensional measurement system for prompt gammas. The measurement system, developed in the present study, incorporates a vertically-aligned one-dimensional array of gamma sensors, a parallel multi-hole collimator, a precision movement system, and a digitizer- and LabVIEW-based automatic data acquisition system. A 45-MeV proton beam of 0.5 nA was delivered to a polymethyl methacrylate (PMMA) phantom, and the two-dimensional prompt-gamma distribution was measured using the developed system. The proton beam range could be quantitatively determined to within a 1.6-mm error by sigmoidal curve-fitting with the Boltzmann equation. A comparison of the prompt-gamma distribution as measured by our detection system with the proton dose distribution as measured independently by using Gafchromic EBT films positioned inside the PMMA phantom showed good agreement. Both results imply that it is, indeed, possible to confirm the patient's proton dose distribution by using two-dimensional prompt-gamma measurements.
