Deformable abdominal phantom for the validation of real‐time image guidance and deformable dose accumulation

Abstract Purpose End‐to‐end testing with quality assurance (QA) phantoms for deformable dose accumulation and real‐time image‐guided radiotherapy (IGRT) has recently been recommended by American Association of Physicists in Medicine (AAPM) Task Groups 132 and 76. The goal of this work was to develop a deformable abdominal phantom containing a deformable three‐dimensional dosimeter that could provide robust testing of these systems. Methods The deformable abdominal phantom was fabricated from polyvinyl chloride plastisol and phantom motion was simulated with a programmable motion stage and plunger. A deformable normoxic polyacrylamide gel (nPAG) dosimeter was incorporated into the phantom apparatus to represent a liver tumor. Dosimeter data were acquired using magnetic resonance imaging (MRI). Static measurements were compared to planned dose distributions. Static and dynamic deformations were used to simulate inter‐ and intrafractional motion in the phantom and measurements were compared to baseline measurements. Results The statically irradiated dosimeters matched the planned dose distribution with an average γ pass rates of 97.0 ± 0.5% and 97.5 ± 0.2% for 3%/5 mm and 5%/5 mm criteria, respectively. Static deformations caused measured dose distribution shifts toward the phantom plunger. During the dynamic deformation experiment, the dosimeter that utilized beam gating showed an improvement in the γ pass rate compared to the dosimeter that did not. Conclusions A deformable abdominal phantom apparatus which incorporates a deformable nPAG dosimeter was developed to test real‐time IGRT systems and deformable dose accumulation algorithms. This apparatus was used to benchmark simple static irradiations in which it was found that measurements match well to the planned distributions. Deformable dose accumulation could be tested by directly measuring the shifts and blurring of the target dose due to interfractional organ deformation and motion. Dosimetric improvements were achieved from the motion management during intrafractional motion.


| INTRODUCTION
Patient motion can reduce the precision of external beam radiotherapy (EBRT), resulting in decreased target coverage and irradiation of nearby healthy structures. This motion can be especially detrimental in the thoracic and abdominal regions of the body, where translational motion and deformation can be on the order of several centimeters. [1][2][3][4] Image-guided radiation therapy (IGRT) has improved the precision of radiotherapy by using different imaging systems to reduce interfractional and intrafractional motion uncertainties. Interfractional motion is monitored during treatment using daily patient imaging data and deformable dose accumulation algorithms. By applying deformable image registration (DIR), the dose each day is deformed to the original patient imaging data that were used to plan the initial treatment. 5 Daily dose maps are used to estimate the delivered cumulative dose distribution during the course of the treatment and assist clinicians in making informed decisions about the possible adaptations to a patient treatment course. 6,7 Furthermore, real-time motion management techniques have been developed to account for intrafractional motion during each fraction resulting from respiratory motion, cardiac motion, peristalsis, etc. During delivery, the treatment is adapted to account for the motion of the target by either gating or tracking the treatment beam. 8,9 Numerous imaging modalities have been utilized to monitor target motion including optical surface tracking, 10,11 magnetic resonance imaging (MRI) guidance, 12,13 ultrasound guidance, [14][15][16] and fluoroscopy. 17,18 Both interfractional and intrafractional motion management strategies have led to the reduction of treatment margins used for a variety of tumor types seen clinically. 19 Clinical implementation of motion management systems requires that they first be validated through measurement. American Association of Physicists in Medicine Task Groups 76 and 132, which discuss respiratory motion management and image registration algorithms in radiotherapy, both recommend end-to-end testing of these systems using quality assurance (QA) phantoms. 20,21 Robust testing of these systems requires a phantom that is able to simulate translational motion and deformation, be compatible with a variety of imaging modalities, and be reusable for multiple experiments. Ideally, these phantoms would also provide three-dimensional (3D) dosimetry, and steps toward the clinical implementation of deformable 3D dosimetry have been made. DEFGEL is a deformable normoxic polyacrylamide gel (nPAG) dosimeter contained within a latex membrane initially proposed Yeo et al. that has been proven by previous work to be well suited for 3D deformable dosimetry. 22 The dosimeter can be irradiated in a deformed state and read out in its original, undeformed state, allowing for the comparison of deformed dose calculations by deformable dose accumulation algorithms to physical measurement. Furthermore, this deformability allows for realistic dynamic deformation of a target during the test of real-time IGRT systems. 23 Other deformable dosimeters have been developed specifically for the testing of deformable dose accumulation algorithms, some examples being Presage-Def, FlexyDos3D, and the incorporation of normoxic methacrylic acid gel (nMAG) in low-density polyethylene containers (LDPE). [24][25][26] However, none of these dosimeters have been incorporated into fully deformable phantoms.
To the best of our knowledge, no phantom exists that includes all of these aforementioned features required to adequately provide robust testing of IGRT systems. For example, some commercial phantoms incorporate rigid motion during treatment but do not to mimic both the translational motion and deformation of the human body. This is especially crucial in testing deformable dose accumulation algorithms since both the rigid image registration and DIR of an algorithm must be validated. Additionally, commercial motion phantoms are limited in the scope of compatible imaging modalities. The purpose of this work was to develop a deformable anthropomorphic phantom that incorporates deformable 3D dosimetry and is compatible with a variety of imaging modalities including both MRI and ultrasound. This phantom and dosimeter pairing was tested for its ability to perform the static deformation measurements beneficial for future testing of deformable dose accumulation algorithms and the dynamic deformation measurements required for future testing of IGRT motion management systems.

2.A | Deformable abdominal phantom development
The phantom apparatus illustrated in Fig. 1 features a deformable plastic abdominal phantom housed in an acrylic shell mounted on an acrylic board. The phantom motion and deformation is driven by a programmable motion stage and plunger incident upon the abdominal section of the apparatus highlighted in yellow in Fig. 1 dosimeter to represent a liver tumor, which will be discussed further in following sections. An example axial slice of a CT of the second phantom is shown in Fig. 2.
The programmable motion stage utilized in this apparatus was the surrogate-axis motion stage of the Washington University 4D Phantom. 32 The motion stage was fully programmable and can be

2.B | Deformable 3D dosimeter
A DEFGEL nPAG dosimeter encased in a PVCP shell was paired with the abdominal phantom. The DEFGEL was fabricated using the materials, heating methods, and mixing techniques described by Yeo et al. 22 The gel was injected into deformable PVCP shells.
The PVCP shells were fabricated using an acrylic outer ring and a 3D-printed insert to create the shape of the inner cavity   formed to its original shape. All dosimeters were MR imaged and analyzed after irradiation using the procedure described for the static irradiation benchmarking.

2.E | Dynamic deformation test
The second phantom apparatus was used to study the effects of dynamic motion and deformation during irradiation and the influence of beam gating on a treatment delivery. During irradiation, the apparatus was set to dynamically move and deform while measuring dose for the testing of real-time IGRT systems. Due to the gating limitations of the Clinac 21EX Linac research linac used for this work, the VMAT treatments that were originally planned to represent liver treatments were replanned and delivered as IMRT plans. Delivery QA was performed using the same method as the static irradiation test, and resulted in a γ pass rate of 100.0% for 3%/3 mm criteria.
Three gel dosimeters were fabricated for this investigation following the aforementioned methods.
The use of motion and beam gating was varied to investigate their effects on treatment delivery. A test treatment was delivered to one of the dosimeters without any motion during the treatment and was used as a baseline and calibration case.

3.A | Static irradiation test
An example of an axial slice of the planned dose distribution and measured distribution in one of the gel dosimeters is shown in Each dosimeter irradiated with the liver SBRT plan was compared to its planned dose distribution using 3D γ-analysis 38,39 with 3%/5 mm and 5%/5 mm criteria. The 5 mm distance-to-agreement (DTA) criterion was chosen so that the DTA criterion would be larger than the largest dimension of the dose map voxel size, 3 mm.
The 5% dose difference was used to account for the 5% estimated k = 1 uncertainty of the gel dosimeter. 40 Pass rates were calculated for the full volume and the central slice with a dose threshold of 20% of the maximum planned dose on both the measured and planned dose maps to remove low-dose regions and regions of oxygen inhibition within the dosimeter and planned low-dose regions.
The average pass rates and standard deviations for each analysis are shown in Table 1.  Table 2 were calculated with a dose threshold meant to include points where either the baseline dose map or the deformed dosimeter dose map displays a dose over 20%.

3.C | Dynamic deformation test
The central coronal slices of the isodose maps of each dosimeter are shown in Fig. 10.   Fig. 12, and the pass rates are shown in Table 3. γ pass rates were calculated using the same 20% threshold as the static deformation experiment. have a tissue-like appearance. 28 These properties of the PVCP material allow for a more versatile phantom that can aid in the testing of a variety of systems that utilize imaging as a motion management method. Also, the addition of the fully programmable motion-driving system allows for the use of realistic motion traces during treatment including simple static deformations.

| DISCUSSION
The pairing of a deformable nPAG dosimeter with the deformable phantom apparatus allows for robust testing of different Static deformation created by the motion stage plunger is consistent with shifts toward deformation sites shown by previous work in deformable 3D dosimetry. 22 In the case of the dosimeters irradiated with a single fraction, the deformation resulted in distinct shifting in the higher isodose levels. Specifically, the plunger pushed the medial side of the dosimeter (the right-hand side of Fig. 7), resulting in the section of gel closer to the plunger to be pushed into the beam, while a section of the gel further from the plunger was pushed out of the beam. When the gel is returned to its original shape at which the treatment was planned at, this push into and out of the irradiated region appears as a shift of the dose distribution approximately equal to the magnitude of deformation on the cranial edge of the phantom, as shown by the profiles in Fig. 8. The dosimeter irradiated with three fractions showed the effects of multiple deformations over the course of treatment, which caused a blurring of the cumulative dose over a large volume. This blurring can be observed in  high-dose profile. Although the central position of the dosimeter had its motion accounted for with beam gating, a widening of the highdose region was still observed. This was likely due to the fact that the dosimeter was still being dynamically deformed during treatment and had a 30% duty factor for its gating window. While the treatment beam was gated on, the dosimeter was still slightly deformed by the phantom plunger compressing the dosimeter toward the trea-

| CONCLUSIONS
A deformable abdominal phantom was developed to contain a removable, deformable nPAG dosimeter to represent a liver tumor within a deformable anthropomorphic phantom. Measurements with the phantom and dosimeter where initially benchmarked with static irradiations measurements that matched well between the dose distributions planned with Eclipse TM and the nPAG dosimeter. The ability to measure the effects of static deformations during treatment and compare these measurements back to a baseline undeformed case was then demonstrated with the phantom. This allows for the phantom apparatus to be used to provide direct measurements of the effects of these deformations and could be used for comparison with calculations made by deformable dose accumulation algorithms in the future. The phantom apparatus was also used to quantify the effects of intrafractional motion and potential improvements of treatment delivery due to the incorporation of beam gating with the intent being to provide a robust test to ensure a system provides a quantifiable benefit to treatment delivery among different real-time IGRT systems. The phantom apparatus developed during this work shows great potential to provide an excellent method for testing and improving the systems currently being developed to monitor and manage patient motion during radiotherapy and improve treatment delivery.