In-House Developed Dose Verification Phantom and Film Analysis Software for Gamma Knife Radiosurgery

Abstract

Objective

The aim of the study was to design a simple and quick plan verification system for Gamma Knife radiosurgery (GKSRS).

Materials and Methods

The latest Icon model of GKSRS includes cone beam computed tomography (CBCT) imaging. An in-house-developed phantom consists of an outer cylinder of 16-cm diameter and a circular nylon plate with threaded groove upon which an inner acrylic cylinder was fixed. This inner cylinder, further divided into two equal parts, holds EBT3 radiochromic film (Ashland Advanced Materials, NJ, United States) in the midportion for irradiation. The phantom was mounted on the Leksell stereotactic frame and set up for CBCT imaging in the treatment couch. The CBCT images were imported into the treatment planning system (TPS) for planning. A prescription dose of 2.5 Gy at the 50% isodose line was planned and a phantom with film was set in the GKSRS couch for irradiation. An in-house-developed MATLAB software (MathWorks, United States) calibrated the film, converted optical density to dose map and performed a comparison with the TPS dose. Film calibration with various doses of 0.5, 1.5, 3, and 5.5 Gy was performed with the same phantom setup as well. All films were scanned 24 hours postirradiation using an EPSON12000XL scanner at 48-bit resolution, 300 dpi, and saved as TIFF images.

Results

Gamma index analysis showed 90.5% pixel pass rate for dose difference (DD) and distance to agreement (DTA) of 3%/2-mm pass criteria. This study demonstrates a quick way to verify the TPS dose distribution during commissioning.

Conclusion

The newly designed apparatus is an effective end-to-end test dosimetry tool for GKSRS, specifically during clinically commissioning.

 

MissingLinkText

Keywords

film dosimetry - Gamma Knife - head phantom - gamma index analysis

Introduction

The Gamma Knife (GK) system is a well-established treatment modality for small brain lesion irradiation that has been in clinical use for several decades. The latest Icon model of GK has the capability to deliver fractionated treatment or multisession irradiation for larger brain lesions using the thermoplastic mask–based hypofractionation methodology. During commissioning of the GK system, vendors provide basic dosimetric tools to perform quality assurance (QA) tests and procedures, which are essential for obtaining clearances from radiation safety regulatory boards for new clinical startup. Yet the commercial dosimetric tools have some limitations and are restricted to specific tests. To overcome such limitations, it is desired for the end user to develop their own QA methodology and tools to validate the newly started clinical facility. The American Association of Physicists in Medicine (AAPM) task group TG-178 mentions that no manufacturer of GK devices provides a comprehensive end-to-end (E2E) test.[1] In this study, we demonstrate an in-house-developed phantom as a dosimetric tool to validate GK radiation delivery.

 

MissingLinkText

Materials and Methods

Recently, our institute clinically commissioned the GK Icon model for patient treatment. This GK model is an updated version to the previous Perfexion model of GK. It has cone beam computed tomography (CBCT) and an intrafraction motion management system, which are some of the features that differentiate it from the previous Perfexion model.[2] The treatment planning system (TPS) used in GK is known as the Leksell Gamma Plan (LGP). The latest inverse planning software is also known as the Leksell Gamma Knife Lightning (LGP Version 11.1). It provides an optimizer that calculates inverse plans based on a set of planning constraints defined by the user.[3]

[Fig. 1A] shows the Leksell stereotactic frame (LF) compatible cylindrical phantom designed in-house for treatment validation purposes. It is a hollow phantom that is different from the commercial solid/water equivalent phantom for GK dosimetry. The phantom assembly consists of an outer cylinder made of acrylic material. Two inner film holder cylinders were designed, each with a diameter of 7.7 cm and a height of 7.6 cm: one for sagittal and coronal irradiation ([Fig. 1B]) and one cylinder for irradiation in axial direction ([Fig. 1C]). The inner cylindrical film holder for performing 2D film dosimetry is the main part of the phantom. It is further subdivided into two parts. For axial irradiation, the inner cylinder is divided into two halves at the mid-transverse plane ([Fig. 1C]). For sagittal/coronal irradiation, the division is along the long axis ([Fig. 1B]). The midportion of the two inner cylinders shown in [Fig. 1B, C] consists of an internal slot for placing the Gafchromic film (ISP, Ashland, United States) for dosimetry.[4] The inner cylinder assembly is mounted on a threaded slot of a nylon circular plate. The whole assembly is further fixed to the LF through a rectangular Lucite plate encased beneath the nylon circular plate. The plate has slots at four corners to mount through metallic screws to the LF to immobilize the phantom during the dosimetry procedure.

Fig. 1 (A) In-house developed phantom for end-to-end (E2E) plan verification. (B) The phantom and its accessories with the film positioned for irradiation in sagittal direction. (C) Film insert prepared for film placement in axial direction.

The phantom with LF was mounted in the couch through a frame adapter for CBCT imaging. The acquired CBCT image of the phantom was imported into the LGP TPS for planning. The stereotactic coordinates were determined by CBCT. A Gafchromic film was placed in the inner cylinder slot to determine the irradiating position. [Fig. 2] shows the phantom positioned in the GK couch for irradiation. A CBCT image was acquired to generate planning based on the phantom images and the stereotactic coordinates obtained.

Fig. 2 Phantom positioned in the GK Icon couch for CBCT imaging.

EBT3 films were analyzed by scanning all the exposed films on a high-quality EPSON Expression12000XL scanner calibrated for radiochromic film analysis. The film strips were scanned in 48-bit resolution in RGB mode, 300 dots per inch (dpi) in transmission mode, and saved in the TIFF format for analysis. No image correction was applied and scanned with consistent orientation.[5]

A graphical user interface (GUI) was created in-house using MATLAB software (MathWorks, United States) to perform film dosimetry. The GUI “FilmQAsoftware” has the option to load films with various dose exposures and perform calibration using inbuilt mathematical curve fit equations ([Fig. 3]). Another GUI “com2iso36” designed in-house with MATLAB was used for gamma index evaluation.[5] [6]

Fig. 3 Film analysis software used to calibrate film measurement and convert to dose maps for analysis.

 

MissingLinkText

Results

[Fig. 4] shows the CBCT image of the phantom. The visible black line in the midportion is the slot area where the Gafchromic film is positioned for irradiation. [Fig. 5] displays the image of the film and irradiation pattern created for E2E verification. The green line is the 50% isodose line displayed for three-shot plan ([Fig. 4]). In the same phantom's mid-region, various film strips are placed for calibration. [Fig. 5A–E] depicts the film irradiation to various calibration doses at 0, 0.5, 1, 3, 5.5 Gy. [Fig. 5F] represents the treatment plan irradiation based on the plan generated in [Fig. 4C].

Fig. 4 (A) CBCT image of phantom (axial view). (B) Midportion of the inner cylinder showing the region where the film was positioned for irradiation experiments (coronal view). (C) Treatment plan irradiation done in the Leksell Gamma Plan TPS and dose distribution (sagittal view). The green line represents the 50% isodose line.

Fig. 5 Film dose exposures: (A) 0 Gy, (B) 0.5 Gy, (C) 1 Gy, (D) 3 Gy, (E) 5.5 Gy, and (F) treatment plan irradiation.

[Fig. 6] displays the status of the Gafchromic film before irradiation and after CBCT imaging. There is no significant transformation of the film after CBCT exposure. There is no significant color change in the Gafchromic film following CBCT imaging. The active component in the EBT3 film forms a blue-colored polymer with the absorption maxima at approximately 633 nm.

Fig. 6 (A) Unirradiated Gafchromic film. (B) Gafchromic film after CBCT imaging.

A second-degree polynomial fit equation converted the optical density (OD) of the treatment irradiated film to a 2D dose map ([Fig. 7]). The calibration plot shows the second-degree polynomial fit between OD and dose (Gy). The R ² value was found to be 0.9954. The vertical error bars in the plot represent the standard deviation. An additional uncertainty in the dose calibration process was noted as the calibration dose was increased. The film irradiated to higher doses tends to saturate in OD due to reduced postirradiation coloration as the dose is increased ([Fig. 7]). The nonlinear trend of the film irradiated at higher dose is an inherent characteristic of the EBT3 film.

Fig. 7 Calibration plot between optical density (OD) and dose (Gy).

[Fig. 8] illustrates the dose profile plotted for various calibration doses and for CBCT imaging dose. The dose value at the center of the profile was found to be up to a 1% variation between the calculated dose and the film-measured dose. Note that CBCT image dose accounts to about 0.047 Gy in EBT3 film when scanned with a Computed Tomography Dose Index (CTDI) of 6.5 mGy.

Fig. 8 Line profile of calibration film doses using second-degree polynomial fit for converting optical density to dose.

[Fig. 9] depicts the post film irradiation exposure of the film placed in the phantom for the treatment setup. The greenish yellow color noted in the phantom is the original film substrate color. The dark band (blue in color) is the region in the film where irradiation occurred.

Fig. 9 Phantom displays the film irradiation. The dark patterns in the film illustrate the GK dose delivery.

[Fig. 10] displays the isodose line overlying between the film measured and the TPS-calculated dose. The innermost blue circle represents the 80% isodose line and the outermost red line represents the 10% isodose line. A close agreement in isodose overlay was found between the TPS dose and the film dose. Gamma pass criterion of 2 mm/3% DTA/DD was set as the pass criterion. The pixel pass rate was around 90.5% ([Fig. 11]).

Fig. 10 EBT3 film versus TPS dose gamma comparison in MATLAB software GUI “com2iso36.”

Fig. 11 (A) Gamma map comparison done in com2iso36 GUI. Gamma pass criteria of 2 mm/3% DTA/DD. (B) Histogram showing pixel pass rate of around 90.5%.

[Fig. 11] represents the gamma distribution map, which is a pixel-to-pixel matching plot between the TPS dose and the film dose. Any gamma value less than 1 indicates close agreement between the TPS dose and the film-measured dose distributions. In [Fig. 11], it is noted that most of the pixels pass the threshold of gamma value less than 1 (represented in the blue color).

 

MissingLinkText

Discussion

The purpose of the newly developed phantom design is to facilitate an easy setup for dosimetry and at the same time demonstrate precision and accuracy in radiation dose measurement. The existing commercial phantoms and devices for GK QA and dosimetry are bulky, complex in design, and limited to specific QA purposes.[7] The commercial phantoms and previously in-house-designed phantoms shown in various studies are not easily adaptable to be immobilized in LF, which defines the stereotactic coordinates for target localization within the skull.[7] [8]

A recent study by Wu et al demonstrated a 3D-printed phantom for routine accuracy check for GK Icon high-definition motion management system.[9] This phantom, however, is not suitable for LF-based dosimetry. Therefore, an E2E dosimetry apparatus for Gamma Knife radiosurgery (GKSRS) verification is required. The newly designed phantom highlighted in this study overcomes these limitations. It is lightweight, simple, and of lower cost. The setup time for each film irradiation in a commercial phantom is 5 to 7 minutes per irradiation, whereas this newly designed phantom reduces this setup time by 50%. It means the setup time for calibration of five films to various doses and one film for treatment irradiation is nearly around 40 minutes in commercial phantoms. Our phantom reduces this experimental setup time to 15 minutes. Please note that the couch travel time and film irradiation time are not taken into account. Irrespective of the phantom used, the couch travel time is nearly 1 minute for each irradiation setup. This fastens the dosimetry procedure in a busy clinic. Another improvisation in the new QA methodology shown our study is that imaging for film localization is done with CBCT instead of routine CT imaging. As far as film dosimetry is concerned, locating the position of the film in the phantom is necessary for planning. The CBCT image of the phantom identifies the film position and directly provides the stereotactic coordinates for shot placement. In addition, the CBCT dose is significantly lesser than the CT imaging dose.

TG-178 states that E2E is generally required for clinical trial accreditation and as a standard practice for other forms of stereotactic radiosurgery. Previous studies mentioned that for a robust QA program to ensure safe use of GKSRS, the manufacturer-required QA checks together with additional user-defined checks are of utmost importance.[1] [2]

After irradiation, the films were processed in “FilmQAsoftware” for calibration and the treatment irradiation film was converted from the OD values to a dose map based on the calibration plot. The GUI “com2iso” built for comparing the TPS-calculated dose and the film-measured dose is shown in [Fig. 10]. The GUI-imported dose maps from TPS and film measurement were matched based on the pixel size, orientation, and dose scaling. Isodose line overlay, gamma index evaluation, etc., were performed.[7] The isodose overlay and gamma index evaluation results of the “com2iso” were validated with commercial film software “VeriSoft Version 7.2” (PTW Freiberg, Germany) as well.[10] The results were in close agreement with commercial software, which showed a pixel pass rate of 88.0%. [Fig. 12] shows the dose distribution of film irradiation and TPS calculation and histogram plot and gamma map in VeriSoft software.

Fig. 12 Gamma map comparison done in VeriSoft software. Gamma pass criteria of 2 mm/3% DTA/DD. The histogram shows a pixel pass rate of around 88.0%.

Earlier Park et al assessed the delivery accuracy of GKRS with the gamma evaluation method using planning dose distribution and film measurement data.[6] They demonstrated the evaluation method using a solid water phantom. A limitation of the study is that the solid water phantom setup for treatment irradiation and calibration for QA testing is a time-consuming procedure. This phantom needs to be mounted in a specialized metallic adapter before setting up in the treatment couch for QA. The punch holes on the film exhibit high OD and a white background, which obscures the pixel intensity during the film scanning procedure.[6] Two previous studies demonstrated EBT2 film verification of fractionated treatment planning with the GK extend system, a relocatable frame system for multiple-fraction or serial multiple-session radiosurgery. This GK system is now considered redundant.[7] [8] In our present study, a simple and effective method for E2E verification has been shown.

Factors Influencing Results

The film dosimetry procedure was followed as per the AAPM task group guidelines.[11] Yet we hypothesize the following as some of the factors influencing film irradiation results:

• Film lifetime: It is recommended to use Gafchromic film within 1 year of its manufacturing period. A change in OD of less than 5 × 10⁻³ per 1,000 lux-day indicates the film stability in light (EBT3 specification manual). Films used for dosimetry in this study were well within the recommended time period.

• Film resolution: Film resolution, set in dpi, needs to be chosen appropriately during film scanning for accurate results. In this study, the resolution was set at 300 dpi during the scanning procedure.

• Postirradiation scanning time: Postirradiation time of at least 24 hours is required for the irradiated film to get fully polymerized before scanning. Scanning was performed 1 day after the irradiation time.

• Calibration: The calibration equation is used to convert the OD map to the dose map. Several authors have proposed mathematical fitting equations to convert film OD to dose. It is better to use the best curve fitting equation for accurate dose prediction depending on the film's response to the calibration dose. The film's R ² value was close to 0.995 at the second-degree polynomial fit calibration. There was no more improvisation in the pixel pass rate when other recommended fitting equations were used.

• TPS dose: Selecting proper TPS 2D dose data to match the film dose is crucial. The TPS dose data are 3D volumetric data. In other words, a stack of 2D dose data out of which exact single 2D data matching with the film dose data needs comparison. Even the slightest variation in film orientation while irradiating and scanning may vary the dose information extracted and eventually the results. Adequate caution was taken in this study to orient the film during irradiation with placement of fiducials and markers ([Fig. 5]). The software “com2iso” offers options to maneuver the orientation of the TPS dose map and film dose map using shift, flip, and rotation features, as shown in [Fig. 10].

• Role of CBCT: The influence of CBCT dose during imaging phantom with film was negligible in our study, but it should not be ignored. The CBCT image dose may have contributed to 0.5 to 1% of the difference in the pixel pass rate.

• Phantom material: Using appropriate material as phantom for dosimetry is crucial. We used acrylic as the phantom material. Attenuation properties vary from solid to water phantoms. Acrylic is a recommended phantom material for dosimetry.[12] The difference in attenuation between about 4-cm-thick air column (that exists between the outer shell and the inner Perspex cylinder of the in-house fabricated phantom) and the water/Perspex column might be a crucial factor. This could be significant enough to warrant a separate study involving variations in the phantom's column thickness. However, this difference was not observed using the film dosimetry system. This factor may be a limitation of the film dosimetry procedure adopted in this study.

In this study, we were able to show a pixel pass rate of 90.5%. There is more scope to improvise our results for better pass rate well above 95% in future studies. The E2E methodology adapted in this study is a viable method for performing routine dosimetry for GK.

 

MissingLinkText

 

MissingLinkText

Conclusion

The newly designed apparatus is an effective dosimetry tool useful for radiosurgery QA. We have shown a simple and effective GK treatment validation technique suitable for use during new clinical commissioning for end users.

 

MissingLinkText

 

MissingLinkText

Conflict of Interest

None declared.

Acknowledgments

A patent was granted by IPR, New Delhi, entitled “An apparatus for verification of radiosurgery using film dosimetry,” with patent no. 496490. The film comparison software “com2iso36” used in this study was developed in collaboration with Prof. Yoichi Watanabe, Department of Radiation Oncology, University of Minnesota, Minneapolis, United States.

Note

This work was presented as poster discussion at the AAPM 66th Annual Meeting & Exhibition (AAPM2024) held in Los Angeles, California, United States, from July 21 to 25, 2024.

• References

• 1 Petti PL, Rivard MJ, Alvarez PE. et al. Recommendations on the practice of calibration, dosimetry, and quality assurance for gamma stereotactic radiosurgery: report of AAPM Task Group 178. Med Phys 2021; 48 (07) e733-e770
• 2 Sjölund J, Riad S, Hennix M, Nordström H. A linear programming approach to inverse planning in Gamma Knife radiosurgery. Med Phys 2019; 46 (04) 1533-1544
• 3 Low DA, Dempsey JF. Evaluation of the gamma dose distribution comparison method. Med Phys 2003; 30 (09) 2455-2464
• 4 AlDahlawi I, Prasad D, Podgorsak MB. Quality assurance tests for the Gamma Knife^® Icon™ image guidance system. J Appl Clin Med Phys 2018; 19 (05) 573-579
• 5 Gafchromic. GAFCHROMIC™ dosimetry media, type EBT-3. Accessed July 2, 2025 at: http://www.gafchromic.com/documents/EBT3_Specifications.pdf
• 6 Park JH, Han JH, Kim CY. et al. Application of the gamma evaluation method in Gamma Knife film dosimetry. Med Phys 2011; 38 (10) 5778-5787
• 7 Natanasabapathi G, Bisht RK. Verification of Gamma Knife extend system based fractionated treatment planning using EBT2 film. Med Phys 2013; 40 (12) 122104
• 8 Bisht RK, Kale SS, Natanasabapathi G. et al. Verification of Gamma Knife based fractionated radiosurgery with newly developed head-thorax phantom. Rad Measur 2016; 91: 65-74
• 9 Wu C, Radevic MB, Glass JS, Skubic SE. Technical note: a 3D-printed phantom for routine accuracy check of Gamma Knife Icon HDMM system. J Appl Clin Med Phys 2018; 19 (04) 299-301
• 10 Shukla AK, Bhamra HK, Rathore NK. et al. Comparison of 2D and 3D gamma evaluation method in patient specific intensity-modulated radiotherapy quality assurance. Int J Res Med Sci 2023; 11: 1160-1164
• 11 Niroomand-Rad A, Chiu-Tsao ST, Grams MP. et al. Report of AAPM Task Group 235 radiochromic film dosimetry: an update to TG-55. Med Phys 2020; 47 (12) 5986-6025
• 12 Khan FM, Gibbons JP. The Physics of Radiation Therapy. 5th ed.. Philadelphia, PA: Lippincott Williams & Wilkins; 2014: 152-153

Publication History

Article published online:
28 July 2025