Accuracy of using high-energy prompt gamma to verify proton beam range with a Compton camera: A Monte Carlo simulation study
Introduction
Due to the unique physical properties of protons, proton therapy can deliver higher doses to the tumor while reducing doses to the surrounding normal tissues. However, the unique properties make proton therapy sensitive to internal organ motion, patient positioning errors and anatomic changes (Knopf and Lomax, 2013). A reliable method for in-vivo dose and range verification is required. During proton therapy, a large number of gamma rays are produced by proton-induced nuclear reactions with the patient's tissue. These rapidly emitted gamma rays are known as prompt gamma (PG) rays. Several studies showed that there is a strong correlation between the distributions of the proton dose and the PGs (Min et al., 2006, Testa et al., 2010, Zarifi et al., 2017). This indicates that PG imaging can be a useful tool for both proton dose and range verification. Like other imaging modalities, PG imaging is a non-invasive imaging technique that has recently been shown to be a promising tool for proton range verification (Krimmer et al., 2018). In addition, PG rays are emitted almost instantaneously from the decay of the excited nuclei. As a result, PG imaging has no biological washout effect (Moteabbed et al., 2011) and is potentially suitable for real-time tracking of the BP during beam delivery.
Over the last ten years, PG imaging has undergone a rapid development which is still continuing. In addition to continuous improvements in hardware and software, several different imaging methodologies were proposed (Krimmer et al., 2018). Two different collimated cameras, a multi-parallel-slit camera and a knife-edge slit camera, were designed to produce a one-dimensional (1D) projection of the beam path (Krimmer et al., 2015, Roellinghoff et al., 2014, Smeets et al., 2012). Compton cameras that consist of two parallel detectors were used to produce three-dimensional (3D) distribution of the PG rays (Singh, 1983, Todd et al., 1974). Furthermore, new proton range verification techniques, including PG spectroscopy (Verburg and Seco, 2014), PG timing measurements (Golnik et al., 2014), and the PG peak integrals (Krimmer et al., 2017), were proposed. Although there is currently no commercial PG imaging modality available, some PG imaging prototypes are being developed and tested under clinical conditions (Polf et al., 2015, Richter et al., 2016).
Several recent studies reported that high-energy PG rays are well correlated to the Bragg peak (BP) position of the proton beam (Hilaire et al., 2016, Zarifi et al., 2017). In particular, the 6.13 MeV PG emission line which originates from the de-excitation of 16O* nuclei exhibits a distinctively close correlation with the BP (Zarifi et al., 2017). Their simulation results showed that the 6.13 MeV PG rays generated from the 100–200 MeV proton pencil-beam irradiation of a homogeneous (or heterogeneous) phantom could accurately predict the BP position and provide an accuracy to 1 mm under ideal conditions (Zarifi et al., 2017). Recently, our simulation results showed that the reconstruction of the 6.13 MeV PG rays obtained by a two-stage Compton camera yielded poor image quality and induced large errors in BP position estimation (~ 17 mm) (Jan et al., 2018). The main reason is that the yield and detection efficiency of high-energy PG rays are extremely low (Jan et al., 2018, Zarifi et al., 2017). In addition, the performance of a Compton camera is limited by the finite energy and spatial resolutions of the detector system, the neutron-induced noise, incomplete absorption of PGs and Doppler broadening (Mackin et al., 2013).
In this study, we aim to improve the BP position estimation using the 6.13 MeV PG image with different position estimation methods. We also investigate the feasibility of using the 6.92 and 7.12 MeV PG emission lines to predict the BP position. These two PG emission lines exhibit a strong correlation with the BP, but have not been studied. We present the results of Monte Carlo simulations in which a water phantom was irradiated with a proton beam. Because the three characteristic PG rays (i.e. 6.13, 6.92 and 7.12 MeV) originate from the de-excitation of 16O* nuclei, we conduct Monte Carlo simulations that evaluate the effect of tissue oxygen contents on the accuracy of the BP position estimation. Finally, we investigate whether the accuracy of the BP position estimation can be improved by combining signals obtained from the characteristic PG emissions of 6.13, 6.92 and 7.12 MeV.
Section snippets
The principle of Compton camera
As shown in Fig. 1(a), a Compton camera generally consists of two parallel plane detectors, often called scatterer and absorber, working in the coincidence mode (Singh, 1983, Todd et al., 1974). In the scatterer detector, an incident gamma-ray photon undergoing a Compton scattering is deflected through a scattering angle () with respect to its original direction. Then, the scattered photon is absorbed in the absorber detector. According to the Compton formula, the scattering angle of the
Results
Fig. 3 shows the 2D projection images and central slices of the Monte Carlo proton dose, 5.93–6.33 and 6.72–7.32 MeV PG images obtained from the simulation of proton beam irradiation of a water phantom. It seems that the distal local peak position of the central slices, as compared to that of the 2D projection image, is more close to the BP position (i.e. white dash line). In addition, compared to the 5.93–6.33 MeV PG images, the 6.72–7.32 MeV PG images show a closer correlation with the BP.
Discussion
In this study, we evaluated the accuracy of BP position prediction obtained from high-energy PG imaging. The simulation results show that the position estimation method has a large effect on the accuracy of the calculated BP position. Both 2D+LM and CS+LM methods can provide more accurate position predictions than the 2D+GM method. This implies that the position with the LM PG intensity along the distal side of the proton beam is favorable for BP tracking. Moreover, the 6.92 and 7.12 MeV PG
Conclusion
The accuracy of using high-energy PG rays to predict the proton BP position was investigated in this simulation study. It was found that the accuracy, based on the difference between PG ray emission and the BP, can be substantially improved using the proposed position estimation method. However, the accuracy of the BP prediction decreases with decreasing oxygen contents. We also observed that the subtraction of the PG images of 6.13 MeV from those of 6.92 and 7.12 MeV can provide an accurate
Acknowledgements
This work was partially supported by the Research Fund of Chang Gung Memorial Hospital, Taiwan [grant numbers: BMRPF46, CIRPG3F0021-3 and CIRPD1F0011-3], and by the Ministry of Science and Technology, Taiwan, R.O.C. (grant numbers: MOST 106-2221-E-002-235, 105-2221-E-182-001 and 106-2221-E-182-016-MY2).
References (30)
- et al.
Geant4—a simulation toolkit
(2003) - et al.
Recent developments in Geant4
Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip.
(2016) - et al.
Prompt-gamma monitoring in hadrontherapy: a review
Nucl. Instrum. Methods Phys. Res. Sect. A Accel. Spectrom. Detect. Assoc. Equip.
(2018) - et al.
First clinical application of a prompt gamma based in vivo proton range verification system
Radiother. Oncol.
(2016) - et al.
Characterization of prompt gamma-ray emission with respect to the Bragg peak for proton beam range verification: a Monte Carlo study
Phys. Med.
(2017) - et al.
Range assessment in particle therapy based on prompt γ-ray timing measurements
Phys. Med. Biol.
(2014) - et al.
Proton therapy monitoring by Compton imaging: influence of the large energy spectrum of the prompt-γ radiation
Phys. Med. Biol.
(2016) - et al.
Effect of 176Lu background on singles transmission for LSO-based PET cameras
Phys. Med. Biol.
(2002) - et al.
Compton camera and prompt famma ray timing: two methods for in vivo range assessment in proton therapy
Front. Oncol.
(2016) - et al.
Accelerated prompt gamma estimation for clinical proton therapy simulations
Phys. Med. Biol.
(2016)
Use of a LYSO-based Compton camera for prompt gamma range verification in proton therapy
Med. Phys.
PSF reconstruction for Compton-based prompt gamma imaging
Phys. Med. Biol.
GATE: a simulation toolkit for PET and SPECT
Phys. Med. Biol.
Factors influencing the accuracy of beam range estimation in proton therapy using prompt gamma emission
Phys. Med. Biol.
In vivo proton range verification: a review
Phys. Med. Biol.
Cited by (1)
Evaluation of using the Doppler shift effect of prompt gamma for measuring the carbon ion range in vivo for heterogeneous phantoms
2020, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated EquipmentCitation Excerpt :In-beam or in-room PET can potentially reduce the significance of the washout effect, but the related technical difficulty remains [8,9]. PG detection during charged-particle treatment has gained interest in range verification because of its instantaneity [10,11]. PG comes from the de-excitation of the excited nucleus produced in the nuclear interactions with a time scale shorter than any biological process.