For real-time dosimetry in electron beam therapy, an integrated fiber-optic dosimeter (FOD) is developed using a water-equivalent dosimeter probe, four transmitting optical fibers, and a multichannel light-measuring device. The dosimeter probe is composed of two inner sensors, a scintillation sensor and a Cerenkov sensor, and each sensor has two different channels. Accordingly, we measured four separate light signals from each channel in the dosimeter probe, simultaneously, and then obtained the scintillation and Cerenkov signals using a subtraction method. To evaluate the performance of the integrated FOD, we measured the light signals according to the irradiation angle of the electron beam, the depth variation of the solid water phantom, and the electron beam energy. In conclusion, we demonstrated that the pure scintillation and Cerenkov signals obtained by an integrated FOD system based on a subtraction method can be effectively used for calibrating the conditions of high-energy electron beams in radiotherapy.
© 2013 Optical Society of America
In general, high-energy electron beams with energy ranging from 6 to 20 MeV have been used in radiotherapy for treating superficial tumors. Unlike a photon beam used in radiotherapy, the electron beam has a clinically useful characteristic that the depth dose falls off sharply beyond the tumors near the skin surface, and hence it can minimize unnecessary exposure to deeper normal tissues . To perform optimum electron beam therapy using a linear accelerator (LINAC), it is necessary to completely fulfill quality assurance (QA) for accurate measurement and maintenance of the beam conditions and the absorbed dose at a given location prior to conducting radiotherapy. Real-time and remote measurements of the absorbed dose distributions are therefore required to determine exact electron beam parameters with dose uniformity. For dosimetry in therapeutic applications, although a variety of ionization chambers have been commercialized and are commonly used, the conventional ionization chambers have some problems regarding application to radiotherapy dosimetry stemming from their large sensing volume and dosimetric materials not being water equivalent [2–4].
Over the past decade, various fiber-optic dosimeters (FODs) based on scintillators and optical fibers have been developed and reported as promising candidates to accurately measure the dose distribution [5–8]. As a sensing material of a miniature scintillating fiber-optic dosimeter (SFOD), a small-sized organic scintillator that has a water-equivalence is generally used to generate scintillating light having dose information. Since optical fiber is used to guide the scintillating light generated from an organic scintillator induced by high-energy radiation beams, SFODs have many favorable dosimetric characteristics over other conventional dosimeters, such as water-equivalence, good flexibility, and immunity to ambient electromagnetic interference (EMI). Despite these advantages, the basic structure and component of most SFODs can cause two critical problems, such as quenching effect and Cerenkov light [9,10]. First, the quenching effect can be occurred at high stopping power when the fluorescent dopant in organic scintillator is temporarily damaged by high-energy radiation beam and it causes a non-proportionality between the energy loss of charged particle in the scintillator and the resultant scintillating light . Second, Cerenkov light can be produced by the direct action of charged particles having a higher energy than Cerenkov threshold energy (CTE) of the optical fiber and it can interfere with accurate dose measurements as an unwanted light signal [10–12]. Various methods to remove the Cerenkov light have previously been reported to measure the pure scintillation signal without a Cerenkov signal [13–16]. However, because it is one of the light signals generated from the dosimeter probe of an FOD by radiation interactions, the Cerenkov light, as well as scintillating light, can be of utility for dosimetry. In recent years, a Cerenkov fiber-optic dosimeter (CFOD) is developed without any scintillator to measure depth dose distribution for high-energy proton and photon beams [9,10]. Although the CFOD can measure real-time dose without the quenching effect at high stopping power, it is a matter of common knowledge that the optical intensities of the Cerenkov light signals generated in optical fibers are much weaker than those of the scintillating light signals of the SFOD. Accordingly, it is difficult to obtain dose information for low-energy radiation beam because of weak Cerenkov light signal that is susceptible to noise.
In this study, we developed a novel integrated FOD system with a multichannel dosimeter probe, which can supplement each other’s shortcomings of the SFOD and the CFOD, to evaluate the usability of both pure scintillation and Cerenkov signals. All sensitive materials of the dosimeter probe are employed to have a nearly water-equivalent characteristic, thereby avoiding the complicated calibration processes necessitated by material differences between the dosimeter probe and surrounding material. By changing the irradiation conditions, we measured four different light signals, which are transmitted from the dosimeter probe induced by an electron beam, simultaneously and then obtained the pure scintillation and Cerenkov signals using a subtraction method.
2. Materials and Methods
2.1. Fabrication of a dosimeter probe
In developing the integrated FOD system, we fabricated a water-equivalent dosimeter probe, which consists of a scintillation sensor and a Cerenkov sensor, to measure scintillating and Cerenkov light simultaneously. Figure 1 shows the structure of the completed dosimeter probe that can output four different kinds of light signals.
First, we fabricated a scintillation sensor to measure scintillating light induced by a therapeutic electron beam. As a sensing element of the scintillation sensor, a plastic scintillating fiber (PSF; BCF-12, Saint-Gobain Ceramic & Plastics) is selected to produce scintillating light. The PSF has a cylindrical shaped core/clad structure, similar to a general optical fiber. The core of the PSF is synthesized with polystyrene (PS) and fluorescent dopants and the cladding material is polymethylmethacrylate (PMMA). The refractive indices of the core and cladding are 1.60 and 1.49, respectively, and the numerical aperture (NA) is about 0.58. The PSF yields approximately 8,000 photons MeV−1 from an ionizing particle and the peak wavelength of the scintillating light emitted from the PSF is 435 nm. The outer diameter and length of the PSF are 0.5 mm and 1 cm, respectively. In order to transmit the scintillating light from the PSF, we used a step-index multimode plastic optical fiber (POF; SH-2001, Mitsubishi Rayon). The core of the POF is made of PMMA resin with a refractive index of 1.49 and the cladding based on a fluorinated polymer has a refractive index of 1.402; thereby the NA is about 0.5. The outer diameter of the POF is 0.5 mm and the cladding thickness is 0.007 mm. The PSF is connected with optical epoxy (DP-100, 3M) to the distal end of a 49 cm long POF. Meanwhile, to remove Cerenkov light generated in a POF, we added a second background sensor to the scintillation sensor. Without employing a PSF, the background sensor is composed of a POF (SH-2001) with a length of 49 cm and the vacancy is filled with a cylindrical PMMA rod of 1 cm length instead of a PSF to avoid dose measurement errors arising from air-gaps. Here, the background sensor can play an important role for producing and subtracting Cerenkov light to provide a pure scintillation signal. As shown in Fig. 2, mixed light signals including both scintillating and Cerenkov light are produced at CH 3 of the scintillation sensor. On the other hand, in the case of CH 4, it transmits only Cerenkov light generated in the POF, which has the same length as the POF in CH 3, with an external electron beam. Therefore, the pure scintillation signal (CH 3 – CH 4) can be obtained by subtracting the output signal of CH 4 from that of CH 3.
Second, we fabricated a Cerenkov sensor, which is composed of two identical POFs (SH-2001) with different lengths to apply the subtraction method. The POF lengths of CH 1 and CH 2 are 45 and 50 cm, respectively, and thereby the length difference of the two POFs is 5 cm. To obtain real-time dose information, the Cerenkov signals (CH 2 – CH 1) measured at each measuring point also can be used and they are obtained by measuring the intensity difference of the Cerenkov light generated from CHs 1 and 2 in the Cerenkov sensor, as shown in Fig. 2.
Finally, the dosimeter probe is constructed by combining the scintillation and Cerenkov sensors. Both ends of the PSF and all POFs are polished with various type of lapping films in a regular sequence to maximize the transmission efficiency of the scintillating and Cerenkov light reaching the light-measuring device. In addition, the uncoupled ends of the PSF and POF are coated with a thin white reflective tape (BC-642, Saint-Gobain Ceramic & Plastics) based on polytetrafluoroethylene (PTFE) to increase the light collection efficiency. The outer surfaces of each fiber and the entire dosimeter probe are surrounded with black heat-shrink tubes to prevent a cross-talk effect and ambient light noise.
2.2. Experimental setup using an integrated FOD system and a CLINAC
To guide the scintillating and Cerenkov light from the dosimeter probe in the radiotherapy room to the light-measuring device in the control room, we used four transmitting POFs (GH-4001, Mitsubishi Rayon) with a diameter of 1 mm and a length of 25 m. Subminiature-type A (SMA) 905 connectors are installed on both sides of the transmitting POFs and the SMA mating sleeves are fitted to enable connection of the transmitting POFs with the dosimeter probe and the light-measuring device. In order to simultaneously measure four light signals transmitted from the dosimeter probe, a multi-anode photomultiplier tube (MA-PMT; R7600U-03-M4, Hamamatsu Photonics) with four channels (CHs) and four low-noise amplifiers (C7319, Hamamatsu Photonics) was used as a multichannel light-measuring device. To calibrate four CH responses of the MA-PMT and amplifier system, each distal-end of the 25 m transmitting POFs, which were already coupled to pertinent CHs of the MA-PMT, was sequentially connected to a fiber-optic light-emitting diode (LED, IF-E92A, Industrial Fiber Optics) with a peak wavelength of 430 nm. Then, we obtained several relationships between the output voltage of MA-PMT and the light power of blue LED to eliminate the influence of sensitivity differences between the four CHs. Finally, each CH response of the MA-PMT and amplifier system, including an offset voltage and an anode radiant sensitivity, was corrected to be equal using a calibration procedure based on a LabVIEW (NI LabVIEWTM 2012, National Instruments) program and a mathematical relation.
Figure 3 presents the experimental setup employed for evaluating the performance of the integrated FOD system using a clinical LINAC (CLINAC, CLINAC® 21EX, Varian Medical Systems) according to the irradiation angle and different depths of a solid water phantom irradiated by electron beams. To produce scintillating and Cerenkov light having dose information, the distal end of the dosimeter probe was accurately located at the isocenter of a CLINAC using a laser alignment system. The target-to-surface distance (TSD) and the field size were set to 100 cm and 10 x 10 cm2, respectively. The energies of the electron beam with 100 monitor units (MUs) are 6 and 9 MeV and the dose rate is 600 MU min−1. Accordingly, the scintillating and Cerenkov light induced by the electron beams were measured by integrating the output voltages of the MA-PMT during a beam irradiation time of 10 sec.
3. Experimental Results
3.1. Angular dependence of an integrated FOD
First, we measured four light signals transmitted from each CH to obtain the angular dependence of the integrated FOD and the distal-end of the dosimeter probe was located at the isocenter of the free-in-air using the experimental setup shown in Fig. 2(a). It has been reported that the intensity of Cerenkov light is dependent upon the irradiation angle between the radiation beam track and the fiber axis and the angle of maximum Cerenkov emission can be theoretically determined as 47.8° when the energy of an incident electron beam and the refractive index of PMMA as a core material are 6 MeV and 1.49, respectively .
Figure 4 shows the variation of the output signals at each CH according to the irradiation angle of a 6 MeV electron beam. By changing the irradiation angle from 0° to 90°, we measured four light signals at each CH simultaneously and all peak values were measured at an angle of around 45° ~47°. The intensity of the Cerenkov signal (CH 2 - CH 1) (i.e., the output signal of a Cerenkov sensor) generated from the limited length (5 cm) of POF also changed in accordance with the irradiation angle, as shown in Fig. 4(b). However, the pure scintillation signals (CH 3 – CH 4) (i.e., the output signal of a scintillation sensor), which are obtained by a subtraction method, have almost uniform values with a maximum difference of about 3.9%, and thus they show an angle-independent characteristic. Through the experimental results in Fig. 4(b), it is obvious that elimination of Cerenkov light as an unwanted signal is necessary to obtain an accurate scintillation signal having absorbed dose information in radiotherapy. Nevertheless, in particular situations, the Cerenkov signal (CH 2 - CH 1) can be expected to be useful to measure and confirm the incidence or trajectory angle of high-energy electron beams over two measurable angular ranges, from 0° to 45° or from 45° to 90° . In the case of the Cerenkov signal (CH 2 - CH 4), its intensity is too weak to evaluate the angular dependence, because it is a supersubtle light signal that is generated from the region where the length difference of two POFs is just 1 cm. When the irradiation angle is 90°, the intensity of the Cerenkov signal (CH 2 - CH 1) is critically low as compared with that of the pure scintillation signals (CH 3 – CH 4), as shown in Fig. 4(b). Therefore, the Cerenkov light can safely be ignored under the specific condition that the dosimeter probe is exposed to the beam perpendicularly .
3.2. PDD measurement of electron beams
Second, we also obtained the depth dose according to the depth variation of the solid water phantom using the total amount of light signals of the integrated FOD. As shown in Fig. 3(b), a solid water phantom with a depth of 15 cm was used as a stopping material and it was placed at the bottom of the dosimeter probe to provide charge particle equilibrium with respect to the surface of the dosimeter probe. In addition, we used extra solid water phantoms with different thicknesses, which were stacked on the dosimeter probe, for measuring the percentage depth dose (PDD). The dosimeter probe was placed at the center of the field size with 10 x 10 cm2 and irradiated from a direction normal to the upper surface of the solid water phantom while maintaining a TSD of 100 cm. Figure 5 presents the depth dose distribution of the 6 and 9 MeV electron beams delivered by a CLINAC and also shows the differences among the four light output signals simultaneously measured from each CH in the dosimeter probe. The mixed light signals of CH 3 contain the scintillating light from the PSF and the Cerenkov light from the POF. Here, in the case of the Cerenkov light generated in CH 3, its amount is almost the same as that of CH 4, because the two POFs of CHs 3 and 4 have the same irradiated length. Since the Cerenkov light yield is proportional to the irradiated length of POF, the Cerenkov lights from CHs 2 and 4 have corresponding output signals because of their different lengths of POFs irradiated in the field size [9,10]. Cerenkov light is also generated in CH 1 by affecting the penumbra and scattered radiation, although the distal end of the POF is placed at the boundary of the irradiation field. In CH 3, the gradient of the rising curve steeper than that of the general depth dose curve of the electron beam; however, it is possible to correct this using a subtraction method that deducts the Cerenkov light of CH 4 from the light signals of CH 3, as shown in Fig. 4.
In general, the central axis depth dose distribution is characterized by the PDD [8,18], and thus the pure scintillation signals (CH 3 – CH 4) and Cerenkov signals (CH 2 – CH 1) are normalized to the responses measured at the depth of maximum dose (R100). In this study, the PDD values obtained by using the integrated FOD are compared with those of a conventional ionization chamber and the PDD curves for the 6 and 9 MeV electron beams are shown in Fig. 6. Using the pure scintillation signals (CH 3 – CH 4) of the integrated FOD and the response of the ionization chamber, we obtained similar PDD curves and two R100 are found to be around a depth of approximately 12 and 22 mm for the 6 and 9 MeV electron beams, respectively. Also, two depths of 80% depth dose (R80) are normally measured at about one-third of each energy . As seen in Fig. 4, almost no Cerenkov light is produced when a therapeutic electron beam is perpendicularly irradiated on the POF. With increasing depth of the solid water phantom having water-equivalence, however, the Cerenkov light yield sharply increases to R100 due increased electron scattering and fluence and the decreased incident angle, as shown in Fig. 7. When the dosimeter probe is located at around R100, the portion of the electrons arriving at the probe with skewed angle (θ) is much higher than that at the skin surface and thus the intensity of Cerenkov light generated in the dosimeter probe is also increased. For this reason, the PDD Cerenkov curve, which is converted from the Cerenkov signals (CH 2 – CH 1), is quite different from the PDD reference curve obtained by the ionization chamber in the build-up region from the skin surface to the R100; however, it falls off sharply with depth beyond R90 (the depth of 90% depth dose), similar to the general PDD reference curve. Unfortunately, unlike dosimetry applications for proton and photon beams used in radiotherapy [9,10], it is difficult to use Cerenkov light to obtain complete PDD values for therapeutic electron beams. Nevertheless, the PDD Cerenkov curve is capable of measuring the specific dose depth including the practical range (Rp) beyond R100 in clinical practice.
In this study, we developed an integrated FOD system with a multichannel dosimeter probe for electron beam therapy dosimetry. The internal structure of the dosimeter probe, which is composed of scintillation and Cerenkov sensors, simultaneously offers four different light signals induced by an electron beam. Using the four light signals transmitted from each CH in the dosimeter probe, the pure scintillation and Cerenkov signals are obtained by a subtraction method from the scintillation and Cerenkov sensors, respectively, and these signals are useful for assessing the various performances of the integrated FOD.
In order to test the integrated FOD system, first, we measured the output signals at each CH according to the irradiation angle of a 6 MeV electron beam. By rotating the treatment head of CLINAC from 0° to 90°, four light signals from each CH are changed with an angular variation, and the peak value of the Cerenkov signal curve, as the output signal of the Cerenkov sensor, is found at an angle of about 47°, close to the theoretical value of 47.8°. However, the pure scintillation signals, as output signals of the scintillation sensor, have almost uniform values regardless of the irradiation angle. Consequently, if using the scintillation sensor in the dosimeter probe, the angular dependence of the FOD system can be overcome with the pure scintillation signals. Furthermore, the Cerenkov sensor is capable of measuring the incident angle of the electron beam, like a protractor. Second, we obtained the PDD pure scintillation and PDD Cerenkov curves using the integrated FOD for 6 and 9 MeV electron beams. The maximum dose values for 6 and 9 MeV electron beams with a field size of 10 x 10 cm2 are found to be at depths of approximately 12 and 22 mm, respectively. In particular, the PDD pure scintillation and the PDD reference obtained by the ionization chamber are similar at both electron energies. In the case of the PDD Cerenkov curve, however, it is quite different from the PDD reference curve in the build-up region due to the various radiation interactions between elections and the medium, the emission angle of Cerenkov light, and the propagation of Cerenkov light in the optical fiber. In other words, the intensity of Cerenkov light generated from the inserted optical fiber in the solid water phantom depends on an incident angle of electron with different depth. In clinical practice, it is expected that the PDD Cerenkov curve can be used for measuring the specific dose depth beyond the R100 despite that Cerenkov light is difficult to use for obtaining complete PDD values for therapeutic electron beams.
In conclusion, we demonstrated that the pure scintillation and Cerenkov signals obtained by an integrated FOD based on a subtraction method can be effectively used for calibrating the conditions of high-energy electron beams in radiotherapy. Based on the experimental results of this study, it is anticipated that the proposed FOD system can serve as a valuable dosimeter to perform QA for accurate measurement and maintenance of both beam conditions and dose distributions prior to conducting radiotherapy.
This research was supported by National Nuclear R&D Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and future Planning (No. 2013004348). Also this research was supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and future Planning (No. 201245539).
References and links
1. F. M. Khan, The Physics of Radiation Therapy (Williams & Wilkins, 1994), Chap. 14.
2. T. Aoyama, S. Koyama, M. Tsuzaka, and H. Maekoshi, “A depth-dose measuring device using a multichannel scintillating fiber array for electron beam therapy,” Med. Phys. 24(8), 1235–1239 (1997). [CrossRef] [PubMed]
3. M. A. Clift, R. A. Sutton, and D. V. Webb, “Water equivalence of plastic organic scintillators in megavoltage radiotherapy bremsstrahlung beams,” Phys. Med. Biol. 45(7), 1885–1895 (2000). [CrossRef] [PubMed]
5. L. Archambault, A. S. Beddar, L. Gingras, F. Lacroix, R. Roy, and L. Beaulieu, “Water-equivalent dosimeter array for small-field external beam radiotherapy,” Med. Phys. 34(5), 1583–1592 (2007). [CrossRef] [PubMed]
6. A. S. Beddar, “Plastic scintillation dosimetry and its application to radiotherapy,” Radiat. Meas. 41, S124–S133 (2006). [CrossRef]
7. F. Lacroix, L. Archambault, L. Gingras, M. Guillot, A. S. Beddar, and L. Beaulieu, “Clinical prototype of a plastic water-equivalent scintillating fiber dosimeter array for QA applications,” Med. Phys. 35(8), 3682–3690 (2008). [CrossRef] [PubMed]
8. W. J. Yoo, J. Moon, K. W. Jang, K.-T. Han, S. H. Shin, D. Jeon, J.-Y. Park, B. G. Park, and B. Lee, “Integral T-shaped phantom-dosimeter system to measure transverse and longitudinal dose distributions simultaneously for stereotactic radiosurgery dosimetry,” Sensors (Basel) 12(5), 6404–6414 (2012). [CrossRef] [PubMed]
10. K. W. Jang, T. Yagi, C. H. Pyeon, W. J. Yoo, S. H. Shin, C. Jeong, B. J. Min, D. Shin, T. Misawa, and B. Lee, “Application of Cerenkov radiation generated in plastic optical fibers for therapeutic photon beam dosimetry,” J. Biomed. Opt. 18(2), 027001 (2013). [CrossRef] [PubMed]
11. A. S. Beddar, T. R. Mackie, and F. H. Attix, “Cerenkov light generated in optical fibres and other light pipes irradiated by electron beams,” Phys. Med. Biol. 37(4), 925–935 (1992). [CrossRef]
13. A.-M. Frelin, J.-M. Fontbonne, G. Ban, J. Colin, M. Labalme, A. Batalla, A. Isambert, A. Vela, and T. Leroux, “Spectral discrimination of Cerenkov radiation in scintillating dosimeters,” Med. Phys. 32(9), 3000–3006 (2005). [CrossRef] [PubMed]
14. B. Lee, K. W. Jang, D. H. Cho, W. J. Yoo, G. R. Tack, S. C. Chung, S. Kim, and H. Cho, “Measurements and elimination of Cherenkov light in fiber-optic scintillating detector for electron beam therapy dosimetry,” Nucl. Instrum. Methods Phys. Res. A 579(1), 344–348 (2007). [CrossRef]
15. B. Lee, W. J. Yoo, K. W. Jang, D. H. Cho, J. K. Seo, J. Y. Heo, S. H. Shin, J. H. Moon, B. G. Park, Y.-H. Cho, and S. Kim, “Measurements of relative depth doses using fiber-optic radiation sensor and EBT film for brachytherapy dosimetry,” IEEE Trans. Nucl. Sci. 57(3), 1496–1501 (2010). [CrossRef]
16. J. Lambert, Y. Yin, D. R. McKenzie, S. Law, and N. Suchowerska, “Cerenkov-free scintillation dosimetry in external beam radiotherapy with an air core light guide,” Phys. Med. Biol. 53(11), 3071–3080 (2008). [CrossRef] [PubMed]
17. T. Watanabe, M. Babzien, K. Kusche, and Y. Yakimenko, “Beam angle measurement using Cherenkov radiation,” in Proceedings of 2005 Particle Accelerator Conference, 2005, pp. 3742–3744.
18. K.-A. Kim, W. J. Yoo, K. W. Jang, J. Moon, K.-T. Han, D. Jeon, J.-Y. Park, E.-J. Cha, and B. Lee, “Development of a fibre-optic dosemeter to measure the skin dose and percentage depth dose in the build-up region of therapeutic photon beams,” Radiat. Prot. Dosimetry 153(3), 294–299 (2013). [CrossRef] [PubMed]