Clinical practice and evaluation of electronic portal imaging device for VMAT quality assurance

Medical Dosimetry 38 (2013) 35-41

Medical Dosimetry journal homepage: www.meddos.org

Clinical practice and evaluation of electronic portal imaging device for VMAT quality assurance Yen-Cho Huang, M.Sc.,* Chien-Yi Yeh, B.Sc.,† Jih-Hsiang Yeh, M.Sc.,* Ching-Jung Lo, M.Sc.,† Ping-Fang Tsai, M.Sc.,† Chih-Hung Hung, Ph.D., M.D.,† Chieh-Sheng Tsai, M.D.,* and Chen-Yuan Chen, B.Sc.† *Department of Radiation Oncology, Chang Gung Memorial Hospital, Keelung Branch, Keelung City, Taiwan; and †Department of Radiation Oncology, Chang Gung Memorial Hospital, Linkou Branch, Taoyuan County, Taiwan

A R T I C L E

I N F O

Article history: Received 05 January 2012 Accepted 08 May 2012 Keywords: IMAT EPID RapidArc Quality assurance VMAT

A B S T R A C T Volumetric-modulated arc therapy (VMAT) is a novel extension of the intensity-modulated radiation therapy (IMRT) technique, which has brought challenges to dose verification. To perform VMAT pretreatment quality assurance, an electronic portal imaging device (EPID) can be applied. This study’s aim was to evaluate EPID performance for VMAT dose verification. First, dosimetric characteristics of EPID were investigated. Then 10 selected VMAT dose plans were measured by EPID with the rotational method. The overall variation of EPID dosimetric characteristics was within 1.4% for VMAT. The film system serving as a conventional tool for verification showed good agreement both with EPID measurements ([94.1 ⫾ 1.5]% with 3 mm/3% criteria) and treatment planning system (TPS) calculations ([97.4 ⫾ 2.8]% with 3 mm/3% criteria). In addition, EPID measurements for VMAT presented good agreement with TPS calculations ([99.1 ⫾ 0.6]% with 3 mm/3% criteria). The EPID system performed the robustness of potential error findings in TPS calculations and the delivery system. This study demonstrated that an EPID system can be used as a reliable and efficient quality assurance tool for VMAT dose verification. 䉷 2013 American Association of Medical Dosimetrists.

Introduction Introductions of new techniques have improved radiotherapy, but at the expense of treatment complexity. Volumetric-modulated arc therapy (VMAT), a novel extension of intensity-modulated radiation therapy (IMRT), is one these new techniques. This rotational therapy delivers prescribed dose in relatively shorter duration and has better dose conformity, uniformity, and normal organ sparing.1–5 Many dosimetric devices for patient-specific quality assurance (QA), such as MatrixX (IBA Radiation Dynamics, Inc., Edgewood, NY), Arc Check (Sun Nuclear, Inc., Melbourne, FL), and Delta 4 (ScandiDos, Inc., Uppsala, Sweden) have been developed for rotational therapy.5,6 Portal dosimetry (Varian Medical Systems, Palo Alto, CA) is one of those dosimetry verification devices. The current generation of electronic portal imaging device (EPID) is composed of amorphous silicon and semiconductor materials. It is not only an imaging tool for treatment setup verification but is also an implement for dosimetry measurements.7–13 Moreover, EPID is a conve-

Reprint requests to: Yen-Cho Huang, M.Sc., Department of Radiation Oncology, Chang Gung Memorial Hospital, Keelung Branch No.200, Lane 208, Jijin 1st Road, Anle Dist., Keelung City 20445, Taiwan. E-mail: [email protected]

nient dosimetry QA tool with a relatively high resolution of 0.392 mm. GLAaS is an algorithm used to derive absolute dose maps from portal images acquired with EPID. The algorithm was developed originally for pretreatment verification. It is a method to compare dosimetric measurements directly against treatment planning system (TPS) calculations. Van Esch et al. investigated EPID dosimetric characteristics (aS500/IAS2, Varian Medical Systems) and used them for IMRT dosimetry.9 Nicolini et al. also tested a new version of the EPID GLAaS algorithm for machine quality assurance and used it for VMAT fields.14,15 Fogliata et al. analyzed the EPID measurements for VMAT quality assurance in different centers.16 However, the previous publications did not use other independent dosimetric systems to verify GLAaS and TPS calculations. The study was divided into 3 parts. The first was to evaluate dosimetric characteristics of the EPID system (aS1000/IAS3). The tests included dose-response linearity, dose-rate dependence, and field sizes dependence. Both 6- and 10-MV photon energies were studied. In the second part, 10 VMAT plans were acquired by the radiographic film system, EPID static method, and EPID rotational method. Finally, the film measurements and the EPID measurements were analyzed and compared with the TPS calculations.

0958-3947/$ – see front matter Copyright 䉷 2013 American Association of Medical Dosimetrists http://dx.doi.org/10.1016/j.meddos.2012.05.004

36

Y.-C. Huang et al. / Medical Dosimetry 38 (2013) 35-41

Fig. 1. Three patterns for dose reproducibility test. A 10 ⫻ 10-cm2 field, an open treatment field, and an IMRT field. The images were achieved by EPID.

Methods and Materials EPID acquisition description Two C-linac iX linear accelerators (LINACs) (Varian Medical Systems, Inc.) with 120 multileaf collimators (MLCs) were used in this study. VMAT, which is named RapidArc, is available on the LINACs. During VMAT treatments, MLC position, gantry speed, and dose rate are modulated to deliver desirable doses. The EPID system was composed of an image detection unit (IDU20), an image acquisition unit (IAS3), a robotic arm (Exact-arm), and a workstation. The detection area was 40 ⫻ 30 cm2, which contained 1024 ⫻ 768 pixels. Each pixel consisted of a light photodiode with a thin film transistor. The pitch between each pixel was 0.392 mm. The incident x-ray hit the scintillating phosphor screen (Gd2O2S:Tb) and generated optical photons. The photons were absorbed by photodiodes, and signals were read out through an analog-to-digital converter. The signals were converted to absolute dose maps by the dosimetric calibration with GLAaS system on the workstation. The response of the detectors is a linear equation composed of primary and transmitted radiation. The total dose for each pixel is the sum of all segments. GLAaS is a method to calibrate EPID detector into dose rather than to predict the EPID response. The detailed rationale and application were described by Nicolini et al.12,13 Compared with other dosimetric systems, hardware such as developers and phantoms are not required for EPID systems. The system settings in this study were: there was no additional buildup on the top of the cassette and the position of EPID robotic arm was set to 0.0/0.0/0.0 (source-to-detector distance [SDD] ⫽ 100 cm). Dark field calibration, flood field calibration, and dosimetry calibration were required before the first use of the system. The dark field calibration was to correct background noise signal, and flood field calibration was to equalize EPID response through the whole area.14 In dosimetry calibration, a diagonal beam profile measured at 5 cm depth of water was

required for off-axis ratio correction. The calibrations were executed for different LINAC dose rates (range from 100 – 600 monitor units [MU]/min) and for different photon energies (6 –10 MV). In practice, only one calibration mode can be selected for one measurement and each calibration mode was used for a specific dose rate and energy. After calibrations, 1 calibration unit (CU) of portal dose of EPID represented 1 cGy. In other words, raw images of EPID were converted to dose at 5 cm depth of water by GLAaS.

EPID characteristics The dose reproducibility of EPID was evaluated by 3 different patterns, including a 10 ⫻ 10-cm2 field, a 3D conformal field, and an IMRT field with sliding technique (Fig. 1). For each field, 3 measurements were compared with each other and they were acquired on 2 LINACs with the same mode EPID system. The dose linearity of EPID was performed with MU ranging from 10 –200. The average over an area of 20 ⫻ 20 pixels at field center was recorded for different MU. The differences were from the comparison of EPID to an ideal linearity. The dose rate is varied in VMAT. Therefore, different LINAC dose rates (range 100 – 600 MU/min) were studied by delivering 100 MU. In addition, the response of using different EPID calibration modes was evaluated. The EPID field size dependence was investigated by setting different field sizes ranging from 3 ⫻ 3 to 30 ⫻ 30 cm2. The average over an area of 20 ⫻ 20 pixels at field center was normalized to that of a 10 ⫻ 10-cm2 field. The value as a function of equivalent square field size was compared between EPID and 0.6 cm3 farmer-type chamber (Exradin A12) measured at 5 cm depth in water. One picket fence pattern proposed by Ling et al. was selected to test the EPID dosimetric system.17 The pattern was originally one of the quality assurance tests for

Fig. 2. The picket fence pattern for EPID dose verification: (A) TPS calculation; (B) EPID measurement; (C) GAI evaluation.

Y.-C. Huang et al. / Medical Dosimetry 38 (2013) 35-41

37

Table 2 EPID reproducibility test

Table 1 Characteristics of 10 RapidArc plans Patient no.

Site

Volume (mL)

Energy (MV)

Arc numbers*

1 2 3 4 5 6 7 8 9 10

Soft palate Palate (left) Buccal Hypopharynx Lower gum Hypopharynx (right) Esophagus Prostate Pelvis bone Rectum

662.8 198.2 900.4 1019.1 738.0 478.1 550.8 799.4 240.4 975.5

6 6 6 6 6 6 6 10 10 10

2 2 3 2 2 2 3 2 2 2

* Two arcs are 2 full rotations, and 3 arcs are 2.5 full rotations.

the RapidArc delivery system. As shown in Fig. 2, 7 strips were irradiated at the same dose but for different dose rates, gantry ranges, and gantry speeds. Before using this pattern to check the EPID dosimetric system, the VMAT delivery system was performed and verified by following the film measurement procedures of Ling et al.17 Verification of clinical VMAT plans Ten selected VMAT plans (22 arc fields) for specific patient QA were listed in Table 1. There were 6 plans for head and neck and 4 for other sites. Optimizations and dose calculations were done on TPS Eclipse v8.6 (Varian Medical Systems). Anisotropic analytical algorithm (AAA) and 2.5-mm calculation grid size were used for calculations. Because EPID is perpendicular to the beam axis during the rotational therapy, original VMAT plans were converted to static IMRT plans for calculation purpose. This method was proposed by Iori et al.18 By keeping the same MLC positions, MU values, and dose rates as the original plans, verification plans were defined at depth 5 cm in water (source-to-axis distance ⫽ 100 cm). The verification processes, which are based on the radiographic film system, are presented in Fig. 3. As shown in process A, the static EPID measurements were compared with the radiographic film measurements. To confirm the consistency between static verification plans and rotational VMAT, EPID measurements acquired with these 2 methods were compared with each other (process B). The comparisons were also done between TPS calculations and film measurements (process C). Furthermore, the EPID rotational measurements were applied for the TPS calculations verification (process D). All measurements and the TPS calculations were exported to OmniPro-I’mrt v1.6 (IBA Dosimetry, Schwarzenbruck, Germany) verification software for analysis. Radiographic film dosimetry system The radiographic film system used consisted of films (Kodak X-Omat V), an automatic film processor (Konica Minolia SRX-101A), and a film scanner (Vidar DosimetryProTM Advantage). To keep accuracy of the radiographic film dosimetry system, we took precautions as follows: careful control of processing environment (chemical solutions, developer immersion time, and temperature): 15-minute warm-up procedure for film scanner; and background corrections for the scanner lights and sensors. The scan resolution was 71 dpi with a 5 ⫻ 5 pixels average filter. Doses to the films were converted from scanning images on OmniPro-I’mrt software by applying calibration curves. Different calibration curves were created for a specific energy (6 or 10 MV) with the same set and same batch of films. Childress et al. studied variations of optical density (OD) over an 18-month period and found them to be as high as 15%.19 To prevent OD variations over time, calibration films and VMAT measurements were irradiated and processed sequentially. The time period between irradiation and processing was controlled as well.

Max difference* (%)

Reproducibility Single 10 ⫻ 10-cm2 field (100 MU) Single 3D conformal field (150 MU) Single IMRT field (150 MU)

LINAC A

LINAC B

0.2 0.3 0.5

0.2 0.4 0.4

* Calculation for each pixel in 3 measurements.

Gamma agreement index analysis In this study, gamma agreement index (GAI) was used for comparison.20 In general, the criteria were set to be a 3-mm difference in distance and 3% difference in dose. Moreover, strict criteria of 2 mm/2% were also used for comparison. All fields were normalized to the maximum dose in each field (Dmax ⫽ 100%) and the low dose exclusion threshold was set to 20%. The percentage of points falling within the criteria provided very important information for dose verification and comparison.

Results EPID characteristics The reproducibility is reported in Table 2. There was ⬍0.5% difference between 3 repeated measurements for all 3 patterns. A similar result (ⱕ0.4%) was found in another LINAC. The EPID dose-response of linearity variation is displayed in Fig. 4. The maximum value is at low dose (10 MU) and is 8% for 6 MV and 9% for 10 MV. However, the variation between 30 and 200 MU is (0.7 ⫾ 1.0)%. The EPID dose rate dependence test is shown in Fig. 5. Each line presents different dose rate calibration modes of EPID. The response is proportional to LINAC dose rate and the maximum variation overall is 2.5%. One finding is that 400 MU/min calibration has ⬍0.8% variation for LINAC dose rate between 100 and 600 MU/min. The field size output factors (OFs) of EPID and farmer-type chamber are illustrated in Fig. 6. Variations are larger for ⬎15 ⫻ 15-cm2 field size and up to 5.6% for 10 MV 30 ⫻ 30 cm2. Nevertheless, it was noted that the equivalent field sizes of VMAT control points (segments) are ⬍15 ⫻ 15 cm2 primarily, and the mean variation of these field sizes is (0.8 ⫾ 0.9)%. From these results, the EPID system shows an uncertainty of ⫾1.4% by propagating all subsections in quadrature (Table 3). By using the method by Ling et al., the accuracy of MLC position, movement, dose rates, and gantry speeds were verified.17 The MLC position was within the tolerance with the film measurements. Accordingly, the EPID image verification results of specific picket fence pattern are shown in Table 4. Compared with the TPS calculation, the GAI evaluation passing rates with criteria 3 mm/3% are (100.0 ⫾ 0.0)% for 6 MV and (99.7 ⫾ 0.0)% for 10 MV respectively. By using more strict

Fig. 3. Schematic diagram of RapidArc plan verification processes. There were 4 different verification processes (A, B, C, and D). Static verification plan was converted from RapidArc. The conversion plan was kept at the same MLC position, MU, and dose rates as the original RapidArc plan.

38

Y.-C. Huang et al. / Medical Dosimetry 38 (2013) 35-41

EPID static and rotational measurements The comparisons of EPID measurements between static and rotational VMAT plans for 10 patients (process B, Fig. 3) were (97.8 ⫾ 1.8)% mean GAI passing rate with strict criteria 2 mm/2% and (100 ⫾ 0.0) % with criteria of 3 mm/3%. EPID rotational measurements and TPS calculations An example of GAI evaluation of EPID rotational measurement compared with TPS calculation (process D, Fig. 3) is shown in Fig. 8. The red area shows the doses failing the criteria and the white area represents doses passing the criteria. The mean GAI passing rate was equal to (99.1 ⫾ 0.6)% with a criteria of 3 mm/3%. Discussion Fig. 4. EPID dose-response linearity variation.

criteria of 2 mm/2%, the GAI evaluation is (98.8 ⫾ 0.1)% for 6 MV and (96.3 ⫾ 0.4)% for 10 MV. Radiographic film measurements The film measurements were compared with the static EPID measurements (process A, Fig. 3) and TPS calculations (process C, Fig. 3). One example of VMAT field for comparison is shown in Fig. 7. The GAI values for 22 arc fields of 10 patients are summarized in Table 5. From the comparison between films and static EPID measurements, the mean GAI passing rate was (94.1 ⫾ 1.5)% with 3 mm/3% criteria. By contrast, the mean GAI passing rate of the comparison between films and TPS calculations was (97.4 ⫾ 2.8)%.

The reproducibility test confirmed the dosimetry performance of the EPID system, but high variation in low dose (⬍30 MU) was found. The under-response for low dose is a result of ghosting effects caused by charge trapping, which was proposed by McDermott et al.21 They stated that the longer the irradiation time, the smaller the relative deficit. The low dose response with different type of EPID have also been discovered by Van Esch et al. and Winkler et al.9,11 It should be noted that 1 field with ⬍30 MU may give an under-reading. However, clinical VMAT fields have ⬎30 MU and they are in the linear response area. Nicolini et al.’s study has shown no differences of EPID doseresponse between different LINAC dose rates with GAI 3 mm/3% criteria.13 The same result (⬍3% difference) was found in this study. EPID dose-response is affected by both dose rate calibration mode of EPID and dose rate of LINAC. During VMAT delivery, LINAC has a variable dose rate (⬍600 MU/min). EPID over-response may happen if using

Fig. 5. Dose-response of EPID irradiated by 100 MU at different LINAC dose rate with 6 different dose-rate calibration modes of EPID.

Y.-C. Huang et al. / Medical Dosimetry 38 (2013) 35-41

39

Fig. 6. Field size factor differences between EPID and water. The value was normalized to 10 ⫻ 10 cm2.

Table 3 The uncertainties of EPID dosimetric measurements

Uncertainty* (%) Condition

Reproducibility

Linearity

0.5 —

0.7 ⱖ30 MU

Dose-rate dependence

Field size OF

0.8 For 400 MU/min EPID calibration mode

0.8 Equivalent field size ⱕ15 ⫻ 15 cm2

Overall simple error propagation 1.4 —

* The value corresponds to the mean of 20 ⫻ 20 pixels at field center.

Table 4 The picket fence pattern for EPID dose verification GAI value (3 mm/3%)

GAI value (2 mm/2%)

Energy

1

2

3

Mean ⫾ SD

1

2

3

Mean ⫾ SD

6 MV 10 MV

100.0 99.7

100.0 99.7

100.0 99.7

100.0 ⫾ 0.0 99.7 ⫾ 0.0

98.8 96.5

98.7 96.5

98.8 95.8

98.8 ⫾ 0.1 96.3 ⫾ 0.4

Fig. 7. One example of RapidArc field. (A) TPS calculation. (B) Portal dose of EPID measurement. (C) Radiographic film measurement.

40

Y.-C. Huang et al. / Medical Dosimetry 38 (2013) 35-41

Table 5 The passing rates of GAI evaluation from different system measurements GAI value (%) Patient no. 1 2 3 4 5 6 7 8 9 10 Mean ⫾ SD (%)

Process A* (3 mm/3%)

Process B* (3 mm/3%)

Process B* (2 mm/2%)

Process C* (3 mm/3%)

Process D* (3 mm/3%)

92.4 96.4 96.5 93.7 93.5 94.6 94.4 92.8 92.6 94.3 94.1 ⫾ 1.5

100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 ⫾ 0.0

98.3 98.2 99.0 99.8 97.1 97.2 100.0 97.0 93.9 98.0 97.8 ⫾ 1.8

98.6 99.9 99.7 97.8 96.3 98.4 94.6 96.2 95.6 99.5 97.4 ⫾ 2.8

99.8 99.6 99.6 99.2 98.7 99.6 98.8 98.1 99.6 98.6 99.1 ⫾ 0.6

* The process is presented in Fig. 3.

default setting of 600 MU/min EPID calibration mode. Based on this study, choosing 400 MU/min dose rate of EPID calibration mode for VMAT measurement may have less variation and it is close to the average of dose rates of VMAT. The EPID verification method reported in this study uses a simple and direct measurement, which is without computing any specific kernel. There is no extra build-up material on EPID while the detectors are inside with a copper layer and a phosphor layer above. The dosimetric characteristic of EPID is not the same as the dose to water so the behaviors have been studied. The field size factor that included both collimator scatter factor and phantom scatter factor was measured. It is an experimental method for pretreatment verification and the accuracy in field center is within an accuracy of 1.4%. In flood field calibration of EPID, the beam profile is assumed to be perfectly flat. This error is corrected by a diagonal beam profile in calibration process. Other complicated correction methods of EPID off-axis position and arm backscatter for specific field size have been published by

Fig. 8. One example of GAI dosimetric evaluation (3 mm/3%).

several groups, and they can be applied to obtain a more accurate dose (⬍1%).22–25 Nicolini et al. have found that the maximum displacement of Exact-arm (Varian Medical Systems) position was 1.5 mm as a result of the gravity effect.15 We performed a similar test and found the arm position shift during rotation was in different directions for different angles. The maximum displacement was 1.2 mm for Exact-arm in this study. For R-arm (Varian Medical Systems), the displacement was up to 9.1 mm and it should be considered in VMAT verification.26 To maintain positional stability of R-arm during gantry rotation, Iori et al. used a homemade holding device attached to a gantry head to deliver VMAT with EPID.18 In the quality assurance of VMAT delivery system, EPID measurements have shown similar results in the picket fence pattern. In Fig. 2, although the EPID image is a little different from the TPS image, EPID exhibits higher resolution and more information. By using gamma analysis, the difference near the penumbra area in each strip is most likely because of the displacement of arm position from gravity, which can also be found in the film measurements. Thus, it proved that the EPID dosimetric system has the capability for measuring different dose rates, gantry ranges, and gantry speeds. Films have been used as an integral dosimetric tool but many factors can affect the quality. The sensitive response of films to low xrays contamination from segments would cause dosimetric variation. Nevertheless, the processing condition and scanner performance were controlled cautiously to maintain film quality. The films measurements confirmed EPID measurements and TPS calculations. The EPID measurements proved that EPID doseresponses between static and rotational VMAT were similar, which included the error of EPID shift during rotation. Accordingly, it demonstrated that EPID rotational verification plan method could be used for VMAT dose verification. The TPS was commissioned and verified before clinical use. Because dose calculations were done in a water phantom, there was less uncertainty in heterogeneous correction. However, MLC transmission factor and tongue-and-groove effect could not be modeled and calculated precisely. This brought few errors from TPS. By contrast, EPID measurement errors were from EPID uncertainties and systematic errors. The EPID uncertainties are listed in Table 3. The systematic errors included MLC leaf sequencing, collimator angle, beam profile, output variation, etc. To maintain the center accuracy of EPID, the mechanical quality assurance of EPID was performed monthly. By contrast, the stability of EPID response was investigated by Nicolini et al. and the article indicated the satisfactory result with a frequency variable from daily to monthly.13 In this study, the static 10 ⫻ 10-cm2 field constancy check was also within a 0.3% difference over a period of 3 months. Although the EPID system presents good stability, the

Y.-C. Huang et al. / Medical Dosimetry 38 (2013) 35-41

sensitivity of hardware changes stepwise and it may deviate from the expectations.14 The need for the EPID recalibrations, which included dark field, flood field, and dosimetry calibration, are required every 3 months. Because EPID is mounted on the gantry, the gantry sag caused by gravity cannot be detected from EPID images. The mechanical center check and radiation center check of the gantry should be done before VMAT pretreatment verification. The quality assurances for VMAT can be found in the study by Ling et al.17 Conclusion The characteristics and dose response of EPID (aS1000/IAS3) were presented in this study. The dose behaviors of EPID proved the robust and stable dose response for clinical practice. Furthermore, the film measurements verified that EPID application for rotational VMAT is reliable. The goal of patient dose verification is to identify potential errors either in TPS calculations or in delivery systems, and EPID has the capability to do this. The EPID system provides a reliable and efficient tool for VMAT quality assurance.

References 1. Cozzi, L.; Dinshaw, K.A.; Shrivastava, S.K.; et al. A treatment planning study comparing volumetric arc modulation with RapidArc and fixed field IMRT for cervix uteri radiotherapy. Radiother. Oncol. 89:180 –91; 2008. 2. Fogliata, A.; Clivio, A.; Nicolini, G.; et al. Intensity modulation with photons for benign intracranial tumours: A planning comparison of volumetric single arc, helical arc and fixed gantry techniques. Radiother. Oncol. 89:254 – 62; 2008. 3. Verbakel, W.F.; Cuijpers, J.P.; Hoffmans, D.; et al. Volumetric intensity-modulated arc therapy vs. conventional IMRT in head-and-neck cancer: A comparative planning and dosimetric study. Int. J. Radiat. Oncol. Biol. Phys. 74:252–9; 2009. 4. Nicolini, G.; Clivio, A.; Fogliata, A.; et al. Simultaneous integrated boost radiotherapy for bilateral breast a treatment planning and dosimetric comparison for volumetric modulated arc and fixed field intensity modulated therapy. Radiol. Oncol. 4:27; 2009. 5. Rao, M.; Yang, W.; Chen, F.; et al. Comparison of Elekta VMAT with helical tomotherapy and fixed field IMRT: Plan quality, delivery efficiency and accuracy. Med. Phys. 37:1350 –9; 2010. 6. Low, D.A.; Moran, J.M.; Dempsey, J.F.; et al. Dosimetry tools and techniques for IMRT. Med. Phys. 38:1313–51; 2011.

41

7. Bertelsen, A.; Hansen, C.R.; Johansen, J.; et al. Single arc volumetric modulated arc therapy of head and neck cancer. Radiother. Oncol. 95:142–50; 2010. 8. McCurdy, B.M.; Luchka, K.; Pistorius, S. Dosimetric investigation and portal dose image prediction using an amorphous silicon electronic portal imaging device. Med. Phys. 28:911–24; 2001. 9. Van Esch, A.; Depuydt, T.; Huyskens, D.P. The use of an aSi-based EPID for routine absolute dosimetric pre-treatment verification of dynamic IMRT fields. Radiother. Oncol. 71:223–34; 2004. 10. Menon, G.V.; Sloboda, R.S. Quality assurance measurements of a-Si EPID performance. Med. Dosim. 29:11–7; 2004. 11. Winkler, P.; Hefner, A.; Georg, D. Dose-response characteristics of an amorphous silicon EPID. Med. Phys. 32:3095–105; 2005. 12. Nicolini, G.; Fogliata, A.; Vanetti, E. GLAaS: An absolute dose calibration algorithm for an amorphous silicon portal imager. Applications to IMRT verification. Med. Phys. 33:2839 –51; 2006. 13. Nicolini, G.; Fogliata, A.; Vanetti, E.; et al. Testing the GLAaS algorithm for dose measurements on low- and high-energy photon beams using an amorphous silicon portal imager. Med. Phys. 35:464 –72; 2008. 14. Nicolini, G.; Vanetti, E.; Clivio, A.; et al. Testing the portal imager GLAaS algorithm for machine quality assurance. Radiat. Oncol. 3:14; 2008. 15. Nicolini, G.; Vanetti, E.; Clivio, A.; et al. The GLAaS algorithm for portal dosimetry and quality assurance of RapidArc, an intensity modulated rotational therapy. Radiat. Oncol. 3:24; 2008. 16. Fogliata, A.; Clivio, A.; Fenoglietto, P.; et al. Quality assurance of RapidArc in clinical practice using portal dosimetry. Br. J. Radiol. 84:534 – 45; 2011. 17. Ling, C.C.; Zhang, P.; Archambault, Y.; et al. Commissioning and quality assurance of RapidArc radiotherapy delivery system. Int. J. Radiat. Oncol. Biol. Phys. 72:575– 81; 2008. 18. Iori, M.; Cagni, E.; Nahum, A.E.; et al. IMAT-SIM: A new method for the clinical dosimetry of intensity-modulated arc therapy (IMAT). Med. Phys. 34:2759 –73; 2007. 19. Childress, N.L.; Salehpour, M.; Dong, L.; et al. Dosimetric accuracy of Kodak EDR2 film for IMRT verifications. Med. Phys. 32:539 – 48; 2005. 20. Low, D.A.; Harms, W.B.; Mutic, S.; et al. A technique for the quantitative evaluation of dose distributions. Med. Phys. 25:656 – 60; 1998. 21. McDermott, L.N.; Nijsten, S.M.; Sonke, J.J.; et al. Comparison of ghosting effects for three commercial a-Si EPIDs. Med. Phys. 33:2448 –51; 2006. 22. Bailey, D.W.; Kumaraswamy, L.; Podgorsak, M.B. An effective correction algorithm for off-axis portal dosimetry errors. Med. Phys. 36:4089 –183; 2009. 23. Berry, S.L.; Polvorosa, C.S.; Wuu, C.S. A field size specific backscatter correction algorithm for accurate EPID dosimetry. Med. Phys. 37:2425–34; 2010. 24. Greer, P.B. Off-axis dose response characteristics of an amorphous silicon electronic portal imaging device. Med. Phys. 34:3815–24; 2007. 25. Rowshanfarzad, P.; McCurdy, B.M.; Sabet, M.; et al. Measurement and modeling of the effect of support arm backscatter on dosimetry with a Varian EPID. Med. Phys. 37:2269 –79; 2010. 26. Iori, M.; Cagni, E.; Paiusco, M.; et al. Dosimetric verification of IMAT delivery with a conventional EPID system and a commercial portal dose image prediction tool. Med. Phys. 37:377–90; 2010.