|Year : 2020 | Volume
| Issue : 3 | Page : 98-104
Application of computed radiography in the quality assurance of linear accelerators in radiotherapy
Saravanan Kandasamy, Vijayaprabhu Neelakandan, Seenisamy Ramapandian, Mourougan Sinnatamby, Muniyappan Kannan
Department of Radiation Oncology, Regional Cancer Centre, JIPMER, Puducherry, India
|Date of Submission||29-Jul-2020|
|Date of Decision||25-Aug-2020|
|Date of Acceptance||19-Sep-2020|
|Date of Web Publication||26-Nov-2020|
Department of Radiation Oncology, Regional Cancer Centre, JIPMER, Puducherry - 605 006
Source of Support: None, Conflict of Interest: None
Introduction: The two mechanisms, optical and radiation fields, operate individually and independently in a Linear Accelerator and cause changes with respect to each other. The standard method for performing the quality assurance (QA) test in radiotherapy involves the irradiation of a radiographic film. In this study, we made an attempt to examine how we could maximize the benefit from an impending filmless environment in the radiotherapy QA program. Aim: The aim is to study the feasibility of using computed radiographs (CRs) in the radiotherapy QA program. Materials and Methods: In this study, the QA tests were performed in Linear accelerators Clinac 600c and Clinac iX, both from Varian Medical Systems, Palo Alto, CA, which was commissioned during September 2004 and July 2011, respectively, were used. Optical and radiation field congruence, radiation isocenter for the gantry, collimator, and couch rotational axis verification, in two linear accelerators were done using gafchromic (EBT3) films and CRs. The standard Gafchromic® EBT3 film, utilized for routine QA were used. The errors estimated were compared and analyzed. Results: The mean error estimated in the QA with both linear accelerators using both QA tools (CR and Film) ranged between 0.053 mm and 0.069 mm, and the standard deviation was estimated to be within 0.062-0.164 mm. Conclusion: The results infer that the QA done with CR is in good agreement with the film. This study poses new challenges to the researchers, task groups, and the regulatory bodies to estimate the frequency of QAs for the newer and the older machines and also, the onset of frequent QAs, once the machine becomes older.
Keywords: Computed radiograph, linear accelerator, optical and radiation field congruence, quality assurance
|How to cite this article:|
Kandasamy S, Neelakandan V, Ramapandian S, Sinnatamby M, Kannan M. Application of computed radiography in the quality assurance of linear accelerators in radiotherapy. Oncol J India 2020;4:98-104
|How to cite this URL:|
Kandasamy S, Neelakandan V, Ramapandian S, Sinnatamby M, Kannan M. Application of computed radiography in the quality assurance of linear accelerators in radiotherapy. Oncol J India [serial online] 2020 [cited 2021 Apr 10];4:98-104. Available from: https://www.ojionline.org/text.asp?2020/4/3/98/301581
| Introduction|| |
The Quality Assurance (QA) test in Medical Linear Accelerators are carried out to assure that the machine parameters are maintained the same as that of the baseline values acquired during their acceptance and commissioning. The aging of the Medical Linear Accelerator and their components may result in gradual changes in the machine parameters, resulting in deviations from the baseline values, which in turn, delivering a degraded quality of treatment to cancer patients in radiotherapy. The International Electrotechnical Commission,, American Association of Physicists in Medicine,,, and American College of Medical Physics laid out procedures and conditions for acceptance testing and commissioning of radiotherapy equipment. Any deviation from the baseline value can result in substandard treatment to the patients.
One of the primary patient set-up processes during radiotherapy treatment involves the alignment of the optical field cross-hair from the collimator of the linear accelerator with the patient marking. It is thus required that the optical field will be replaced by a radiation field once the radiation beam is “ON.” Especially when image-based patient set-up is not done, this becomes more important. Practically, both the optical and radiation sources in the linear accelerator are from two independent components. As these two mechanisms operate independently, they may cause changes with respect to each other. This could potentially lead to wrong dose delivery. Task Group TG-142 gives the recommendation for coincidence between optical and radiation field to be within 2 mm or 1% per side for symmetric field sizes. The standard method for performing this QA test involves the irradiation of a radiographic film. However, in this study, an attempt to use film-less QA and to enable digital format, which helps to have better archiving and improved image quality, was made. We used a computed radiography (CR) system, which is used for conventional X-ray imaging.
Electronic portal imaging devices with appropriate software have been developed to quantitatively study optical and radiation field congruence.,,,,, Medical X-ray imaging nowadays moves towards filmless environment and uses CR instead of film. Similarly, in radiotherapy for various QA's Gafchromic films are being replaced with various semiconductor-based detectors. Hence, we adopted CR for performing the QA in radiotherapy. In this paper, we report a simple QA test tool using CR in linear accelerators.
For external beam therapy (EBT) dosimetry, radiochromic films are used widely. In 2011, International Specialty Products released a new film generation, hromic EBT3 film, an improved version of EBT2 films to carry out the above QA's. Although the active layer in the EBT2 films, which was used earlier, is the same, the EBT3 has a special polyester substrate that inhibits the creation of Newton ring interference patterns in images obtained using flatbed scanners. While using EBT3 films, postprocessing irradiation is not necessary and is capable of using without the requirement of the darkroom. Other main advantages are its high spatial resolution (at least 25 μm), low energy dependence, and near tissue equivalent (Zeff= 8.73). The measurement and analysis of films are time-consuming and implicates set-up errors, predominantly while marking the beam axis center on the film, that usually lead to the substantial level of uncertainty in the results compared with the 2 mm lenience on overall rotation of gantry, collimator, and couch axis. Our aim of this study is to examine how we could maximize the benefit from the filmless environment.
| Materials and Methods|| |
In this study, two Linear accelerators Clinac iX (Linear Accelerator 1) and Clinac 600c (Linear Accelerator 2), both from Varian Medical Systems, Palo Alto, CA, which were commissioned in 2011 and 2004, respectively, were used. At the time of this study, Clinac 600c had been put in clinical service for nearly 14 years and the other Clinac iX that was relatively new, had completed 7 years. The QA test that required the use of films were performed in both of these two linear accelerators. The standard Gafchromic® EBT3 film, utilized for routine QA, was used. The film is considered to be the golden standard for radiotherapy QA. Carestream directview vita Computed Radiography (CR) systems were used for image detector and analysis. The system consists of the Vita CR scanner and a software package that operates the scanner. An image viewing and archiving software package that supports the DICOM with variable size of phosphor plate was used.
Optical and radiation field congruence
In a linear accelerator, the secondary collimator jaws restrict the beam in 'X' and 'Y' direction using has four jaws X1, X2, Y1, and Y2. The orientation of the jaws with respect to beam's eye view is shown in [Figure 1]. It is important to carry out a periodical check of the movement of these jaws and estimate and correct the errors in the jaw positions with isocentre as the origin of the plane. Prior to that, the displayed values in the control station and mechanical positions of collimator jaws were checked and found to be within tolerance.
|Figure 1: Orientation of jaws in secondary collimator of a linear accelerator|
Click here to view
The CR phosphor plate is placed on the couch of the linear accelerator, and with the guidance of an optical field and the crosshairs proper orientation of the CR is achieved. The orientation of the CR plate itself enables it to determine the gantry side and relocate the secondary collimator jaws (X1, X2, Y1, and Y2) during evaluation with the image. Calibrated front pointer and optical distance indicator are used to set the surface at 99.5 cm, so as to keep the sensitive plane of the phosphor plate at isocenter at 100 cm. The set-up is shown in [Figure 2]a and [Figure 2]b.
|Figure 2: Setup for optical and radiation field congruence (a and b), and image acquired using gafchromic film (c) and Computed radiography (d)|
Click here to view
In the linear accelerator, the X-ray energy of 6 MV and a dose rate of 300 MU/min are selected, with both the collimator and couch positioned at 0°. The selected field size in the collimator is 10 cm x 10 cm. A spherical radio-opaque marker with a dimension of around 2 mm was kept on the surface of the CR at the optical central axis (crosshair). The orientation of the CR is noted to locate the gantry position in the image; only then it will be possible to identify the X1, X2, Y1 and Y2 jaws perfectly. Exposure for 2 MU are delivered with energy of 6 MV X-rays at 300 MU/Min dose rate. It is not necessary to place a bolus over CR as is required for the films since the CR plate is sandwiched between intensifying screens, which provides the necessary build-up.
Similarly, the same procedure is repeated for the Gafchromic® EBT3 film. The film is placed on the couch of the linear accelerator, by using an adhesive tape, the film was firmly fixed to the couch. The optical field and the cross-hair were used to maintain the proper orientation of the film. Similarly, to identify the gantry side, a lead marker is kept over the film on the gantry side. It is mandatory to ensure that the optical field covers within the film. A prick in the film enables to identify the optical crosshair even after removing the film from the setup position. A bolus of 1 cm thickness is to be used for build up. This is done to eliminate the perturbing influence of the incident electron contamination. The gantry, collimator, and couch are all kept at 0°. Using a calibrated front pointer, the film is placed at isocenter, and the bolus surface is kept at 99 cm so that the film is kept at 100 cm. An exposure for 200 MU was delivered with the energy of 6 MV photons, at 300 MU/Min dose rate. “One has to take note of the MUs used for film and CR.” As CRs are based on speed imaging system 2 MU is sufficient to create the latent image, to that of 200 MU required for films. The optical field is marked with markers and compared with the gray shade formed due to exposure of radiation in the film. The images acquired by film and CR are given in [Figure 2]c and [Figure 2]d. There is a huge difference in the MU used between CR and film. This is due to the fast and slow imaging properties of CR and Gafchromic® EBT3, respectively. The fast film captures image rapidly reducing movement artifacts and delivers a very low dose to patients. The slow film captures the image with higher dose delivery over a prolonged time but provides a very good resolution. Hence, only fast films are used in imaging for the patients, but for QA and other research and development works, we can use slow films, which gives higher resolution. The uniqueness of this study is to use the fast imaging system in QA, and hence require lesser MU that results in minimum time to complete the test.
Coincidence of radiation isocenter with collimator axis rotation
The set-up is shown in [Figure 2]a and [Figure 2]b, and similar to the previous procedure, the field size in the collimator was maintained at 0.2 cm × 40 cm, only a slit of opening was given to obtain the star pattern in the image. In telecobalt unit, if the field cannot be reduced to 0.2 cm, one can use shielding blocks to achieve the minimum fields. The collimator is rotated to various angles 0°, 45°, 90°, and 315°. This angle selection was made so that the collimator covers the maximum rotation angle. Exposure was given in all these angles. For CR plate an exposure of 2 MU and for Gafchromic® EBT3 film exposure of 200 MU are delivered. The images acquired by film and CR are shown in [Figure 3]a and [Figure 3]b. The same process is repeated with different gantry angles of 0°, 90°, 180°, and 270° and correspondingly, the orientation of the detectors are changed to face the primary radiation field at normal angle always.
|Figure 3: Image acquired in collimator axis rotation using film (a) and Computed radiography (b)|
Click here to view
Coincidence of radiation isocenter with couch axis rotation
The experimental setup for radiation coincidence with couch axis rotation is similar to the setup of radiation isocenter with collimator axis. With the collimator at 0°, the couch is rotated to various angles 45°, 135°, 0°, and 270° as shown in [Figure 4]a; these angle selections are made so that the couch covers the maximum possible rotation angle. The couch has a restriction of rotation, its movement is limited by a semi-circle between 90° and 270°; hence, the couch angle rotation is different from the gantry and collimator rotation angles. With 6 MV photons and 300 MU/min, exposures are given in all the aforementioned angles. For CR plate, an exposure of 2 MU and for Gafchromic® EBT3 film an exposure of 200 MU are given. The image acquired by film and CR is shown in [Figure 4]b and [Figure 4]c.
|Figure 4: Setup for couch axis rotation (a), image acquired with gafchromic film (b) and Computed radiography (c)|
Click here to view
Coincidence of radiation isocenter with gantry axis rotation
The orientation of CR/Gafchromic® EBT3 are kept perpendicular to the couch surface; care should be taken that the plane of the CR and Gafchromic® EBT3 are exactly perpendicular to the gantry rotational axis of the linear accelerator. The experimental setup, for the coincidence of radiation isocenter with gantry axis rotation, is displayed in [Figure 5]a and [Figure 5]b. The CR cassettes can be kept perpendicular to the couch, but for film the slab phantoms can be used to render necessary support. The collimator field size is kept at 0.2 cm × 40 cm. The gantry is rotated to various angles 45°, 315°, 180°, and 270°. Exposures are given in all the aforesaid angles for both CR and Gafchromic® EBT3 with the same beam parameters used for previous QA tests. The image acquired by CR is shown in [Figure 5]c.
|Figure 5: Setup for gantry axis rotation (a and b), image acquired with Computed radiography (c)|
Click here to view
The gray shade created in the Gafchromic films are used to make the measurements. To identify the center of the field, the prick made with a needle gives the necessary assistance; this is done with the guidance of the optical crosshairs. Measurements are made on the film using a calibrated ruler. The blades of the secondary collimator are measured from the pricked center to the edges of the gray shade. Beams Eye view of secondary collimator jaws at collimator 0° is shown in [Figure 1]. Similarly, for the collimator, couch, and gantry rotational axis, the margin between the inner circle and outer circle are estimated from the star image pattern and using a well-calibrated protractor, the angles are estimated. To avoid interpersonal variation during image analysis with the CR, we fixed a standard window leveling in imaging software.
| Results|| |
The independent variables are the data from QA tests using CR and gafchromic film, obtained from two different linear accelerators. The outcome variables are:
- Deviation in optical and radiation field alignment with respect of four jaws
(X1, X2, Y1, and Y2)
- Mechanical and Radiation gantry axis deviation
- Mechanical and Radiation collimator axis deviation
- Mechanical and Radiation couch axis deviation.
The errors estimated with various QA tests carried out using Film and CR are given in [Table 1], [Table 2], [Table 3], respectively. Descriptive statistical data in [Table 4] and [Table 5] gives a clear idea that CRs are as good as using films. The mean error estimated in the QA program with both linear accelerators using both QA tools (CRs and Films) are found to be within 0.053–0.069 mm, and the standard deviation was estimated to be within 0.062–0.164 mm.
|Table 1: Estimated error in linear accelerator 1 and 2 using gafchromic film for secondary collimator jaws|
Click here to view
|Table 2: Estimated error in linear accelerator 1 and 2 using computer radiograph for secondary collimator jaws|
Click here to view
|Table 3: Estimated error in linear accelerator 1 and 2 using gafchromic film and computed radiograph for collimator, gantry and couch axis coincidence|
Click here to view
[Figure 6] and [Figure 7] gives the error distribution for Linear Accelerator 1 and 2, respectively. [Table 1] and [Table 2] reveals that there are more nil errors in Linear Accelerator 1 compared to Linear Accelerator 2. This is an expected outcome as Linear Accelerator 2 is comparatively an older machine, and the error expected is more. This also gives an inference that the protocol for carrying out the QA test and its frequency should be modified with respect to the age of the machine.
|Figure 6: Error distribution in Linear Accelerator 1 for optical field congruence|
Click here to view
|Figure 7: Error distribution in Linear Accelerator 2 for optical field congruence|
Click here to view
Due to technical reasons and age of the machine, the couch rotation in Linear Accelerator 2 is not possible and hence that particular QA is excluded from this study.
| Discussion|| |
Patel et al. conducted a study on the use of the Kodak 2000RT system with large Agfa CR plates as a substitute of film for linear accelerator and simulator QA. They suggested that the CR images may be used as a good alternative to radiographic film for a widespread range of QA for megavoltage and kilovoltage X-ray fields. They found it easy to use, giving standard results consistently, correction with data's were clearly introduced and examined in commercial dosimetry software; RIT114.
Kerns and Anand performed a similar study and developed a method to rapidly and precisely quantify optical and radiation field coincidence using CR photo-stimulable phosphor plates. In addition, they also accurately located the crosshair by which the optical and radiation portals are spatially related to one another, letting clinical physicists to gather meaningful results during the performance of monthly QA. They have cautioned that the CR system may saturate, and necessary precautions are to be taken to avoid it. In our study, we have adopted the same technique in estimating the other additional QAs like verification of collimator, couch, and gantry axis rotation.
Sheridan and McNulty in their study regarding the image quality of CR versus digital radiography (DR), had concluded that CR system with the integrated exposure fluence was superior to the DR system in terms of recognizing clear, unleaded, annealed glass foreign body association. They also accomplished that DR has endured many improvements in recent years involving improved dynamic range and digital enrichment and is now a commendable competitor for the latest technologies. In this study, we have come out with similar outcomes. Apart from the routine use for imaging, CR has been identified as a QA tool in radiotherapy.
Njeh et al. studied the design for swift and simple QA of the optical and radiation field using electronic portal image device (EPID) or computed radiography (CR) and determined that either an EPID or CR provide a modest and brisk method to verify optical and radiation field congruence. They have also introduced deliberate errors of various magnitudes and tracks, and concluded that the results are, as good as that of the films. In our study we have gone a step ahead to do a few additional QA tests using CR.
The two-dimensional array devices are the substitutes for the optical-radiation field coincidence with precision, but inaccuracy can exist. Such precision by using the devices can be enhanced substantially. However, it requires analysis of the results and verification of accuracy, which initiates intra and inter-observer disparity that confines the accuracy of these approaches.
A wide variety of programmatic and system-level factors along with an effectual QA program is a key component of quality improvement and patient safety toward the successful radiotherapy treatment. A suboptimal treatment delivery without proper supervision at each step of the planning and delivery course can lead to poor clinical outcomes, severe patient injury, and mortality.,
It has been demonstrated that it is feasible to use CR effectively in carrying out few QAs in radiotherapy. The image quality is quite superior to make the necessary assessment of the QA tests. Further study is needed to vary the detector speed property, by which CR can also be used to detect the variation of photon fluence.
The method presented here is expedient for numerous reasons. First, it does not rely on the observer for interpretation. However, while using films to identify the field borders i.e., the penumbra region, the field borders are arbitrary and vary from person to person. However, while using CR, the field borders were identified with sub-millimeter accuracy. CR method is swift; hence in any busy clinic with a CR already commissioned, this can be adopted with ease. It is inexpensive, reusable, with the additional advantage of proper archiving. However, the limitation with the sensitive area of the CR restricts for performing the QA with bigger field size, and the saturation property of CR restricts its usage over a long period.
The errors estimated with CRs and Gafchromic® EBT3 films shows good agreement, while CR gives better consistency and reproducibility than films. The option of archiving in CR enables to maintain proper documental evidence of the QA. As the CR plates are reusable, it is cost-effective. This method is simple, reproducible and better than the conventional method compared to the robust assessment using Gafchromic® EBT3 film with no observer interpretation or uncertainty needed. The set-up time of CR is comparatively lesser than Gafchromic® EBT3 films and based on our results, we recommend its use as a standard QA tool for radiotherapy QAs. In addition, it is readily available in almost all clinics and radiotherapy centers. The major setback in estimating the error from Gafchromic film is inter and intrapersonal variation due to the poor image quality in the penumbra region. As the films have a stipulated shelf life, the image quality further degrades if used beyond a particular period. The availability of a properly calibrated ruler is another set back in most of the clinics. These setbacks were overcome significantly while using the CR systems.
| Conclusion|| |
It has been demonstrated that CRs can be effectively used in performing various QAs in radiotherapy. Although there are specific tolerance values given for various QA in radiotherapy, there are no guidelines given specifically on the frequency in which they are to be carried out against the age of the linear accelerator. In our institute, we have two Linear Accelerators of which one is 14 years old and the other 7 years old. The recommendations on QA for both these linacs are the same, but based on our results, the frequency on which the QA has to be carried out need not be the same. This study poses new challenges to the researchers, task groups, and the regulatory bodies to estimate the frequency of QA tests on the newer and older machines and the onset of frequent QAs and come out with new recommendations. It is essential to carry QAs more frequently in older machines compared to newer ones. This study also gives a promising scope for the feasibility of using CR in HDR Brachytherapy QAs as well.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
International Electrotechnical Commission. Particular Requirements for the Safety of Gamma Beam Therapy Equipment. Vol. 2. European Committee for Standards - Electrical: International Electrotechnical Commission; 1987. p. 1-64.
International Electrotechnical Commission. Medical Electron Accelerators in the Range of 1 MeV to 50 MeV Guidelines for Functional Performance Characteristics. Vol. 1. European Committee for Standards - Electrical: IEC 60977; 1989. p. 1-52.
Nath R, Biggs PJ, Bova FJ, Ling CC, Purdy JA, van de Geijn J, et al
. AAPM code of practice for radiotherapy accelerators: Report of AAPM Radiation Therapy Task Group No. 45. Med Phys 1994;21:1093-121.
Svensson GK, Baily NA, Loevinger R, Morton RJ, Moyer RF, Purdy JA, et al
. Physical aspects of quality assurance in radiation therapy, AAPM Report #13. Am Inst Phys 1984;13:5-8.
Das IJ, Cheng CW, Watts RJ, Ahnesjö A, Gibbons J, Li XA, et al
. Accelerator beam data commissioning equipment and procedures: Report of the TG-106 of the Therapy Physics Committee of the AAPM. Med Phys 2008;35:4186-215.
”Radiation Control and Quality Assurance in Radiation Oncology: A Suggested Protocol,” American College of Medical Physics Report Series No. 2. Reston, VA: American College of Medical Physics; 1986.
Klein EE, Hanley J, Bayouth J, Yin FF, Simon W, Dresser S, et al
. Task Group 142 report: Quality assurance of medical accelerators. Med Phys 2009;36:4197-212.
Ho A, Thomadsen B, Paliwal B. On visual interpretation of light localization/radiation field coincidence films. Med Phys 1995;22:237-8.
Dunscombe P, Humphreys S, Leszczynski K. A test tool for the visual verification of light and radiation fields using film or an electronic portal imaging device. Med Phys 1999;26:239-43.
Luchka K, Chen D, Shalev S, Gluhchev G, Rajapakshe R. Assessing radiation and light field congruence with a video based electronic portal imaging device. Med Phys 1996;23:1245-52.
Kirby MC. A multipurpose phantom for use with electronic portal imaging devices. Phys Med Biol 1995;40:323-34.
Njeh CF, Caroprese B, Desai P. A simple quality assurance test tool for the visual verification of light and radiation field congruent using electronic portal images device and computed radiography. Radiat Oncol 2012;7:49.
Prisciandaro JI, Herman MG, Kruse JJ. Utilizing an electronic portal imaging device to monitor light and radiation field congruence. J Appl Clin Med Phys 2003;4:315-20.
Manikandan A, Sarkar B, Nandy M, Sureka CS, Gossman MS, Sujatha N, et al
. Detector system dose verification comparisons for arc therapy: Couch vs. gantry mount. J Appl Clin Med Phys 2014;15:4495.
Casanova Borca V, Pasquino M, Russo G, Grosso P, Cante D, Sciacero P, et al
. Dosimetric characterization and use of GAFCHROMIC EBT3 film for IMRT dose verification. J Appl Clin Med Phys 2013;14:4111.
IS. Products Advanced and Materials. Gafchromic EBT3 Scan Handling Guide; 2012.
Patel I, Natarajan T, Hassan SS, Kirby MC. The use of computed radiography for routine linear accelerator and simulator quality control. Br J Radiol 2009;82:827-38.
Kerns JR, Anand A. The use of computed radiography plates to determine light and radiation field coincidence. Med Phys 2013;40:111707.
Sheridan N, McNulty JP. Response to letter re: Computed radiography versus indirect digital radiography for the detection of glass soft-tissue foreign bodies. Radiography (Lond) 2017;23:82.
Manikandan A, Sarkar B, Holla R, Vivek TR, Sujatha N. Quality assurance of dynamic parameters in volumetric modulated arc therapy. Br J Radiol 2012;85:1002-10.
Nath R, Anderson LL, Meli JA, Olch AJ, Stitt JA, Williamson JF. Code of practice for brachytherapy physics: Report of the AAPM Radiation Therapy Committee Task Group No. 56. American Association of Physicists in Medicine. Med Phys 1997;24:1557-98.
Rivard MJ, Coursey BM, DeWerd LA, Hanson WF, Huq MS, Ibbott GS, et al
. Update of AAPM Task Group No. 43 Report: A revised AAPM protocol for brachytherapy dose calculations. Med Phys 2004;31:633-74.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]