A novel and fast method for proton range verification using a step wedge and 2D scintillator

Transcription

1 A novel and fast method for proton range verification using a step wedge and 2D scintillator Jiajian Shen, a) Bryce C. Allred, Daniel G. Robertson, Wei Liu, and Terence T. Sio Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, AZ 85054, USA Nicholas B. Remmes Department of Radiation Oncology, Mayo Clinic Rochester, Rochester, MN 55905, USA Sameer R. Keole and Martin Bues Department of Radiation Oncology, Mayo Clinic Arizona, Phoenix, AZ 85054, USA (Received 5 May 2017; revised 11 June 2017; accepted for publication 22 June 2017; published 31 July 2017) Purpose: To implement and evaluate a novel and fast method for proton range verification by using a planar scintillator and step wedge. Methods: A homogenous proton pencil beam plan with 35 energies was designed and delivered to a 2D flat scintillator with a step wedge. The measurement was repeated 15 times (3 different days, 5 times per day). The scintillator image was smoothed, the Bragg peak and distal fall off regions were fitted by an analytical equation, and the proton range was calculated using simple trigonometry. The accuracy of this method was verified by comparing the measured ranges to those obtained using an ionization chamber and a scanning water tank, the gold standard. The reproducibility was evaluated by comparing the ranges over 15 repeated measurements. The sensitivity was evaluated by delivering to same beam to the system with a film inserted under the wedge. Results: The range accuracy of all 35 proton energies measured over 3 days was within 0.2 mm. The reproducibility in 15 repeated measurements for all 35 proton ranges was mm. The sensitivity to range variation is 0.1 mm for the worst case. This efficient procedure permits measurement of 35 proton ranges in less than 3 min. The automated data processing produces results immediately. The setup of this system took less than 5 min. The time saving by this new method is about two orders of magnitude when compared with the time for water tank range measurements. Conclusions: A novel method using a scintillator with a step wedge to measure the proton range was implemented and evaluated. This novel method is fast and sensitive, and the proton range measured by this method was accurate and highly reproducible American Association of Physicists in Medicine [ proton pencil beam scanning, proton range measurements, quality assurance, scintilla- Key words: tor, wedge 1. INTRODUCTION The interest in using protons for radiation therapy is rapidly increasing due to beneficial integral dose characteristics. 1 As compared to conventional x-ray radiation therapy, the sharp distal dose fall-off of protons enables significantly lower integral dose and the capability of improved sparing of healthy tissue, while still maintaining a conformal dose distribution to the target. 2 On the other hand, the finite range and sharp distal dose fall-off, if not accurately quantified and verified, can lead to significant under-dosing of the target and over-dosing of critical healthy tissue or organs at risk. 3,4 Therefore, proton range quality assurance (QA) is essential to ensure high-quality proton radiation therapy. The gold standard to quantify proton range is to measure the depth dose with an ionization chamber in a water tank. However, a water tank is typically used only during initial machine commissioning and annual QA. It is an inconvenient tool for frequent QA testing such as daily, weekly and monthly measurements. Alternative methods to quantify proton range exist but have other disadvantages. For example, Zebra (IBA Dosimetry, Schwarzenbruck, Germany), a commercial multiple layer ionization chamber device, has been used to measure proton ranges. However, its maximum measurement uncertainty was reported to be more than 1 mm. 5 Gafchromic EBT Film has been used for proton range measurements during commissioning, and a 0.5-mm accuracy was reported. 6 However, film is not efficient for a frequent constancy check of ranges for multiple proton energies because of postprocessing requirements and the number of films required. Volumetric scintillator detectors have also been used for fast range verification in proton therapy, 7 and a few recent studies reported that sub-millimeter accuracy was achieved, 8,9 but this is not currently a commercially available technology. In this letter, we propose a novel analytic method using a large 2D scintillator of 30 cm 9 40 cm combined with a customized wedge device to efficiently measure the proton range to 0.2 mm precision. Notably, the large 2D scintillator 4409 Med. Phys. 44 (9), September /2017/44(9)/4409/ American Association of Physicists in Medicine 4409

2 4410 Shen et al.: Fast proton range check by 2D scintillator 4410 device also allows for simultaneous quality assurance for spot size and spot position. 10,11 Other technologies discussed above do not currently offer such efficient and convenient allin-one functionality for range, spot size, and spot position QA. 2. METHODS AND MATERIALS 2.A. The 2D scintillator and the step wedge The 2D scintillator XRV4000 (Logos Systems Int l, Scotts Valley, CA, USA) has an effective measurement area of 30 cm 9 40 cm, and spatial resolution of mm/ pixel. XRV4000 has 4.9-mm of acrylic buildup which includes the top acrylic cover, acrylic under the cover for mechanical support of the scintillator material, and the scintillator itself. A sketch of the wedge LCW-200 Plus (Logos Systems Int l, Scotts Valley, CA, USA) is shown in Fig. 1. The wedge was made of acrylic with six graduating steps (named A to F from the highest to the lowest, respectively). On the top of each step is a pair of 45 wedges. The highest step (i.e., step A) is 20 cm tall in the center and 15 cm on both sides. The height decreases by 2 cm for each subsequent step. As a result, step F is 10 cm tall in the center and 5 cm on the sides. The footprint of the wedge is 10 cm (front dimension) 9 15 cm (side dimension). The wedge is attached to a 1.2 cm acrylic base. The combined thickness of the wedge base and the native device buildup is 1.69 cm, as shown in Fig B. The beam design and the range detection A scanning proton beam treatment plan with 35 energy layers was designed with energies from 121 to 189 MeV and beam ranges in water from 10.6 to cm. Each layer was composed of 625 spots, evenly distributed at 6-mm spot spacing. This gave an intended field size of 15 cm 9 15 cm which covered the entire footprint of the step wedge. The spots were equally weighted in order to deliver a laterally homogeneous fluence to the step wedge. As proton spots from a given energy layer are delivered to the step wedge, the range is determined by the water equivalent thickness (WET) of the step wedge through which it traveled. An example of the images obtained from the scintillator by this method for E = MeV (range = 174 mm) is shown in Fig. 2(a). The dark pixels at steps A and B are due to low energy beams unable to reach the scintillator. Two bright regions occur in both steps C and D, where the WET of the wedges is equivalent to the proton range. Hence, the proton range can be derived from the position of the brightest pixels on the scintillator image, by simple trigonometric principles. For steps E and F, the protons were energetic enough to pass completely through the wedges and the scintillator. After raw images were captured by scintillator and the wedge system, an in-house data analysis to derive the proton range was developed. The longer curve in the Fig. 2(b) was obtained by averaging the light intensity profiles in the top rectangular ROI as highlighted in the left of Fig. 2(a). The purpose of the ROI average was to get a smooth Bragg peak curve FIG. 1. The front and side-view sketch of the step wedge, with the 16.9 mm acrylic base and buildup of the scintillator also included. The peak heights of wedges A to F are 200 mm, 180 mm, 160 mm, 140 mm, 120 mm and 100 mm, respectively. The side heights are 50 mm shorter than the corresponding peak heights in a uniform fashion. The footprint is 100 mm mm. The wedge angle is 45.

3 4411 Shen et al.: Fast proton range check by 2D scintillator 4411 FIG. 2. (a): an image captured by the scintillator with the step wedge for protons at E = MeV. The bright pixels in wedges C and D were caused by the Bragg peak dose deposition. The rectangles identify the regions of interest (ROI) for data analysis. (b): the longer curve is the averaged light intensities of the top rectangular ROI as shown in left image. The shorter curves represent the fit to the longer curve. The x labels mark the locations of 80% of the peak intensities, which were used to calculate the half- distance between Bragg peaks (L). [Color figure can be viewed at wileyonlinelibrary.com] by removing the effects of camera readout noise. Equation 1 was used to fit the light intensities (y) of the smoothed Bragg peaks, 2 C2 ðc3 xþ y ¼ C1 exp C4 1 pffiffi 2 1 þ erf C3 x 2 C2 C5 pffiffi (1) 2 C5 C4 where C1 to C5 are the fit parameters, and x is its corresponding position on the scintillator. Equation 1 was drawn from the analytical formula for the proton integral depth dose, 12,13 but it was greatly simplified as we only needed to fit the critical region near the Bragg peak and distal fall-off. After curve fitting was performed, the points with 80% of the peak intensity on the distal fall-off side were identified [see cross labels in Fig. 2(b)]. The half-distance between Bragg peaks [L, as shown in Fig. 2(b)] is defined as half of the distance between the 80% points. 2.C. The range quantification method The proton range is the water equivalent thickness (WET) that the beam traveled between the position where proton pencil beam entered the wedge surface and the position where the same protons generated Bragg peak brightness in the scintillator. Since the proton beam has divergence in both X and Y directions, the distance that the beam travels can be calculated using the three projections in the X, Y, and Z directions, respectively. In Fig. 3, the beam path d was projected to the X plane and Y plane, respectively. Using trigonometry and the Pythagorean Theorem, we can derive, qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi R wedge ¼ C WET d ¼ C WET d 2 x þ d 2 2 y þ d z (2) d z ¼ h a ¼ h SAD x h SAD x L L d x ¼ L a ¼ L SAD x h SAD x L L (2.a) (2.b) d y ¼ b d z (2.c) SAD y d z where C WET is the conversion factor from acrylic thickness to WET, h is the peak height of the wedges which includes the base and the buildup, L is the half peak distance measured by the scintillator as defined above, b is the distance from the isocenter to the center of the wedge that the protons pass through, and SAD x (135 cm) and SAD y (192.5 cm) are the virtual SAD of the beam in the X and Y directions, respectively. The WET of the wedge was measured using the scanning water tank, with a resulting value of C WET ¼ 1: D. Accuracy and reproducibility test The scintillator device (XRV4000) was placed on the couch and the center of the scintillator device was aligned

4 4412 Shen et al.: Fast proton range check by 2D scintillator 4412 FIG. 3. The projections of the beam passing through the wedge into the X and Y planes are represented. The d x,d y, and d z are the projections of the beam trajectory d in the X, Y, and Z directions, respectively. h is the peak height of the wedges which includes the base and the buildup, L is the half peak distance as defined above, b is the distance from the isocenter to the center of the wedge that protons pass through, and SAD x (135 cm) and SAD y (192.5 cm) are the virtual SAD of the beam in the X and Y directions, respectively. [Color figure can be viewed at wileyonlinelibrary.com] to isocenter using the laser. The step wedge was placed on the top of the scintillator device and the center of the wedge was aligned to the laser. The proton beam described in the above section was delivered 5 times a day for 3 days. The data analysis process was performed, and the proton range R wedge was derived. For each day, R wedge from five repeated measurements was averaged, and these average ranges were compared with the ranges measured by a scanning water tank (R 80, 80% of the peak at the distal falloff): ΔR =R wedge R 80. The reproducibility of this novel technique was then evaluated by comparing the R wedge results from all 15 measurements. In order to test the sensitivity to the change of ranges, one EB3 Gafchromic film (Ashland, Bridgewater, NJ, USA) was placed under the wedge and five repeated beams were delivered. The WET of 10 stacked films was measured using the scanning water tank, giving a measured single-film WET of 0.36 mm. R wedge was derived from this data set without accounting for the film, so that the use of the film can be treated as generating a new set of 35 energies whose corresponding ranges were smaller than the original set by 0.36 mm (film WET). R wedge from five repeated measurements with film was averaged and compared with the ranges by scanning water tank. 3. RESULTS 3.A. Range measurement accuracy The accuracy of ranges measured with the step wedge is shown in Fig. 4. Please note that the range measured by step wedge was systematically lower than the scanning water tank measurements (Δ = 0.51 mm). The cross, plus and circle symbols show the range accuracy of all 35 energies measured on days 1 through 3, after the application of a systematic shift of Δ = 0.51 mm. As a result, the accuracy of the range as measured by step wedge was within 0.2 mm after the systematic shift was applied. The standard deviation of the day-to-day variation was very small at only 0.04 mm. The star symbols show the range accuracy measured with the step wedge with a film inserted under the wedge. The star symbols are on average 0.35 mm (r = 0.04 mm) less than green symbols for all 35 energies, and with minimum and maximum deviations at 0.26 mm and 0.42 mm, respectively. 3.B. Range measurement reproducibility The reproducibility of proton beam ranges as measured with the step wedge for all 35 energies in 15 measurements is shown in Fig. 5. The reproducibility is within 0.3 mm, with a standard deviation of mm. The dashed lines mark one and two standard deviations (r), respectively. 4. DISCUSSION The underlying principle of this novel method was to project the depth dose into the lateral direction using a wedge, so that the depth dose could be measured with a flat scintillator placed perpendicular to the beam and imaged by a CCD camera. As shown in Fig. 4, the accuracy of the proton range measured with the scintillator and wedge was 0.2 mm after a systematic shift of 0.51 mm was applied, which is

5 4413 Shen et al.: Fast proton range check by 2D scintillator 4413 FIG. 4. The accuracy of the range measured with the step wedge and scintillator. The cross, plus and circle symbols are the range accuracy measured from day 1 through day 3, after a constant 0.51 mm shift was applied. The star symbols show the range accuracy measured with a 0.36 mm WET film inserted under the wedge. On average, the ranges of star symbols were 0.35 mm (solid line) shorter than those of the cross, plus and circle symbols. [Color figure can be viewed at wileyonlinelibrary.com] FIG. 5. The reproducibility of the range measurements with the scintillator and step wedge, including the entire set of 15 repeated measurements for all 35 energies. The dashed lines mark one and two standard deviations (r = mm), respectively. [Color figure can be viewed at wileyonlinelibrary.com] comparable to the precision of water tank measurements. As the star symbols demonstrate in Fig. 4, the test easily detects a 0.36 mm range change with the average and the worst sensitives at 0.01 mm and 0.1 mm, respectively. While the system precision is comparable to that of water tank measurements, our novel method is much faster, requiring only 5 min to set up the system and 3 min of beam delivery time to measure 35 beam ranges. If the water tank was used instead, it

6 4414 Shen et al.: Fast proton range check by 2D scintillator 4414 would take about 1 h for setup and several hours to measure all of these proton ranges. The setup is simple, fast, and reproducible, requiring only the alignment of the wedge with the setup lasers. The highly reproducible results as shown in Fig. 5 indicate that this method is robust. This method can be translated to the other scintillator, such as Lynx (Ion Beam Applications, SA, Louvain-la-Neuve, Belgium). The 2D scintillator is used in our institute and other institutes for the quality assurance of the spot sizes and spot positions. 10,11 These spot size and position measurements are not included in this paper because they are beyond the scope of this paper. However, another distinct advantage with this novel method is that it allows the proton range, along with the spot size and spot positions, to be measured using a single device. This saves cost by reducing the number of measurement devices required for QA, and it saves time by removing the need to set up multiple detectors for different measurements. A 0.51 mm system shift has to be applied in order to get 0.2 mm accuracy as shown in Fig. 4. The source of the submillimeter systematic shift will be investigated in future experiments. A number of uncertainties may cause this submillimeter deviation, including uncertainties in the acrylic WET conversion factor, the physical thickness of the system, and/or the pixel resolution. The linear energy transfer dependence of the scintillator response, also known as the quenching effect (the scintillator material not having a linear response to radiation), may also play a role in this systematic shift. We selected the distal 80% of the peak intensity to define R wedge, which follows our institution s proton range definition in water (i.e., R 80 ). A previous study has also found the distal 80% to be appropriate for scintillator-based range measurements. 8 However, the quenching effect may change the shape of the distal Bragg peak, leading to differences in the distal 80% location between the scintillation light curve and the actual dose distribution in water. In order to have a more meaningful comparison of the ranges measured by the wedge and the water tank, quenching correction should be applied in the future. 14 The current wedge design can only be used to measure ranges from about 10 to 24 cm. Although it is possible to add buildup to extend this range to higher energies, the current design does not allow measurement of lower energy beams. A future work is to modify the current step wedge, so that ranges from 4 cm to 32.4 cm (as available in our proton scanning machine) can be measured. 5. CONCLUSION A novel method to measure proton range using a 2D scintillator detector and a step wedge was implemented and evaluated. This novel method is fast, robust, sensitive and highly reproducible. The range accuracy is within 0.2 mm, which is comparable to that of a scanning water tank, the gold standard. ACKNOWLEDGMENTS We thank Logos Systems engineer Mr. Brett Nelson for providing specific information on XRV4000. CONFLICT OF INTEREST The authors have no relevant conflicts of interest to disclose. a) Author to whom correspondence should be addressed. Electronic mail: shen.jiajian@mayo.edu. REFERENCES 1. Particle Therapy Co-Operative Group. Particle therapy facilities in operation: Information about technical equipment and patient statistics; pp. Retrieved from 2. Smith AR. Vision 20/20: proton therapy. Med Phys. 2009;36: Lomax AJ. Intensity modulated proton therapy and its sensitivity to treatment uncertainties 1: the potential effects of calculational uncertainties. Phys Med Biol. 2008;53: Paganetti H. Range uncertainties in proton therapy and the role of Monte Carlo simulations. Phys Med Biol. 2012;57:R99 R Dhanesar S, Sahoo N, Kerr M, et al. Quality assurance of proton beams using a multilayer ionization chamber system. Med Phys. 2013;40: Zhao L, Das IJ. Gafchromic EBT film dosimetry in proton beams. Phys Med Biol. 2010;55:N291 N Fukushima Y, Hamada M, Nishio T, Maruyama K. Development of an easy-to-handle range measurement tool using a plastic scintillator for proton beam therapy. Phys Med Biol. 2006;51: Archambault L, Poenisch F, Sahoo N, et al. Verification of proton range, position, and intensity in IMPT with a 3D liquid scintillator detector system. Med Phys. 2012;39: Hui C, Robertson D, Alsanea F, Beddar S. Fast range measurement of spot scanning proton beams using a volumetric liquid scintillator detector. Biomed Phys Eng Express. 2015;1: Lin LY, Ainsley CG, Mertens T, De Wilde O, Talla PT, McDonough JE. A novel technique for measuring the low-dose envelope of pencil-beam scanning spot profiles. Phys Med Biol. 2013;58:N171 N Lin L, Huang S, Kang M, Solberg TD, McDonough JE, Ainsley CG. Technical note: validation of halo modeling for proton pencil beam spot scanning using a quality assurance test pattern. Med Phys. 2015;42: Ulmer W, Schaffner B. Foundation of an analytical proton beamlet model for inclusion in a general proton dose calculation system. Radiat Phys Chem. 2011;80: Shen J, Liu W, Stoker J, et al. An efficient method to determine double Gaussian fluence parameters in the eclipse proton pencil beam model. Med Phys. 2016;43: Robertson D, Mirkovic D, Sahoo N, Beddar S. Quenching correction for volumetric scintillation dosimetry of proton beams. Phys Med Biol. 2013;58: