Clinical commissioning of a new patient positioning system, SyncTraX FX4, for intracranial stereotactic radiotherapy

Abstract Background & Aims A new real‐time tracking radiotherapy (RTRT) system, the SyncTraX FX4 (Shimadzu, Kyoto, Japan), consisting of four X‐ray tubes and four ceiling‐mounted flat panel detectors (FPDs) combined with a linear accelerator, was installed at Uonuma Kikan Hospital (Niigata, Japan) for the first time worldwide. In addition to RTRT, the SyncTraX FX4 system enables bony structure‐based patient verification. Here we provide the first report of this system's clinical commissioning for intracranial stereotactic radiotherapy (SRT). Materials & Methods A total of five tests were performed for the commissioning: evaluations of (1) the system's image quality; (2) the imaging and treatment coordinate coincidence; and (3) the localization accuracy of cone‐beam computed tomography (CBCT) and SyncTraX FX4; (4) the measurement of air kerma; (5) an end‐to‐end test. Results & Discussion The tests revealed the following. (1) All image quality evaluation items satisfied each acceptable criterion in all FPDs. (2) The maximum offsets among the centers were ≤0.40 mm in all combinations of the FPD and X‐ray tubes (preset). (3) The isocenter localization discrepancies between CBCT and preset #3 in the SyncTraX FX4 system were 0.29 ± 0.084 mm for anterior‐posterior, −0.19 ± 0.13 mm for superior‐inferior, 0.076 ± 0.11 mm for left‐right, −0.11 ± 0.066° for rotation, −0.14 ± 0.064° for pitch, and 0.072±0.058° for roll direction. the Pearson's product‐moment correlation coefficient between the two systems was >0.98 in all directions. (4) The mean air kerma value for preset #3 was 0.11 ± 0.0002 mGy in predefined settings (80 kV, 200 mA, 50 msec). (5) For 16 combinations of gantry and couch angles, median offset value in all presets was 0.31 mm (range 0.14–0.57 mm). Conclusion Our results demonstrate a competent performance of the SyncTraX FX4 system in terms of the localization accuracy for intracranial SRT.


| INTRODUCTION
Image-guided radiotherapy (IGRT) is becoming crucial for further innovation in conformal radiotherapy, as the use of IGRT ensures that high-precision techniques are delivered as planned. 1 In particular, high localization accuracy (typically within 1 mm) is needed in intracranial stereotactic radiotherapy (SRT) in order to not compromise the local control and to minimize the risk of intracranial complications. 2 Several research groups reported that IGRT techniques including orthogonal kV-imaging, 3 oblique kV-imaging, [4][5][6][7][8][9] kV-conebeam computed tomography (CBCT), 2,5,[9][10][11][12][13] Fig. 1(a). The principle of this system for RTRT is similar to the previous RTRT systems described by Shirato et al. 16 and Shiinoki et al. 17 However, the SyncTraX FX4 differs from the previous systems in two notable ways. 16,17 First, these systems' detectors differ; that is, an image intensifier (I.I.) is used in the previous systems, whereas FPDs are used in the SyncTraX FX4 system. The influence of image distortion caused by the use of an I.I. has thus been eliminated, and bony structure-based verification became possible with the SyncTraX FX4 as a patient verification system (Fig. 2). Second, the designs of the x-ray tube and the detector are different. As shown in Fig. 1(b), the imaging positions of the x-ray tubes and FPDs of the SyncTraX FX4 can be selected from a total of four combinations called "presets". In addition, since the configuration of the x-ray tubes in the SyncTraX FX4 system has changed from the previous rail-type system to the fixed type, it is possible to switch the imaging direction promptly compared to the previous systems. Thus, since there are very few blind angles with regard to the gantry and couch angles of the SyncTraX FX4, this system could be assumed to be effective for intracranial SRT. However, there is no report about positioning verification for radiotherapy using the SyncTraX FX4. We conducted the present study to provide a first report of the clinical commissioning of a SyncTraX FX4 system for intracranial SRT.

2.A | Image quality
For the evaluation of the image quality of each FPD in the Sync-TraX FX4 system, we used an image evaluation phantom (Shimadzu). This phantom was composed of aluminum plates with different resolution and thickness values, and it was attached in front of an FPD before measurement, as shown in Fig. 3 Fig. 4(b)]. For contrast resolution, we calculated the contrast-to-noise ratio (CNR) using the following formula: where S1 mean and S2 mean are the mean pixel values over a region For the evaluation of uniformity, we divided the images taken without attaching the phantom into nine ROIs [ Fig. 4(d)] and calculated the coefficient of variation (CV) using the following formula: where SD M is the mean value of the standard deviation for each pixel and Mean M is the mean value of the average for each pixel.
We set the acceptance criteria as a measurement error <0.5 mm for scaling, a CNR >50 for contrast resolution, and a CV <5% for uniformity.

2.B | Imaging and treatment coordinate coincidence
We evaluated the coincidence between the imaging coordinate and that of the treatment systems based on the TG-142 report. The spatial displacement between the radiation center in the TrueBeam and the image center in the SyncTraX FX4 was measured using a phantom with a tungsten sphere (Fig. 5). The tungsten sphere was placed at the radiation center using RIT 113 ver. 6.3 software (Radiological Imaging Technology, Colorado Springs, CO).
We performed the measurements at gantry angles of 0°, 90°, 180°, and 270°so that the error of each axis was within 0.05 mm.  from pixel unit to mm units. We also evaluated the coordinate coincidence between the kV-planar images obtained with an on-boardimager (OBI) and those obtained with the SyncTraX FX4.

2.C | Comparison of the localization accuracies of CBCT and SyncTraX FX4
We used the head phantom shown in Fig. 6(a) to determine and compare the localization accuracy of CBCT and the SyncTraX FX4. After we shaped a commercial thermoplastic mask (Qfix; Avondale, PA) to immobilize the head phantom, images were taken at a slice thickness of 1 mm using a CT scanner (SOMATOM Definition AS, Siemens, Erlangen, Germany). The CT isocenter was set at the center of the skull base.
Next, the CT dataset was taken into the treatment planning device Eclipse ver. 13 (Varian Medical Systems), and the planning-isocenter for the treatment plan was also set to the center of the skull base.
For the positional verification, we used skin marker-based matching to place the phantom at the planning isocenter. The couch was randomly moved within a range of 0-20 mm for translational shifts (AP, SI, LR) and within a range of 0°-1.5°for rotational shifts around the AP axis (yaw), the SI axis (roll), and the LR axis (pitch). We then carried out bony structure (BS)-based matching using CBCT to correct the setup errors through automatic image registration. Subsequently, BS-based matching was performed automatically using the SyncTraX FX4. A total of 20 datasets of setup errors were acquired for each preset. We examined the correlation of the shifts between the CBCT and SyncTraX FX4 system.
We determined the Pearson's correlation coefficient between the CBCT and SyncTraX FX4 system relative to the position of the skin marker-based matching. We then plotted the differences between each shift against the average shift by performing a Bland-Altman analysis to assess the fixed bias. The average value, standard deviation, and root-mean-square (RMS) of the difference between each shift were calculated for each preset.

2.D | Radiation dose of the SyncTraX FX4
With the SyncTraX FX4 system, imaging x-rays can be delivered at an angle of 37.7°from two x-ray tubes (#1 and #2) with a source-to- The measurement was carried out five times in each of the 40 imaging conditions, and the average value and the standard deviation were calculated. In the present study, under the assumption that the radiation dose of two tubes is equal for each of the four presets, twice the average value was taken as the air kerma per one measurement.

2.E | End-to-end test
As shown in Fig. 6(b), a commercial thermoplastic mask (Qfix) was shaped for a head phantom with a 5-mm-dia. gold marker embedded. Planning CT images were then obtained at a slice thickness of 1 mm using a CT scanner (SOMATOM Definition AS).
The isocenter on the treatment planning was set at the marker

3.A | Image quality
The discrepancy between the actual and measured distances for FPD #3 was within 0.5 mm. For the spatial resolution, a chart of 2.3 lp/mm was identified. The CNR for contrast resolution was 101.9, and the CV for uniformity was 0.16%. The image quality results of the other three FPDs satisfied each of the four criteria, as shown in Table 1.

3.D | Radiation dose of the SyncTraX FX4
The relationship between the tube current and air kerma for each voltage of the x-ray tube for presets #3 and #4 is illustrated in Fig. 9.
In both presets, the air kerma increased with the increase in the tube

3.E | End-to-end test
The results in the end-to-end test acquired by EPID for preset #3 are shown in Fig. 10. For 16 combinations of gantry and couch angles, the median value of the offset between the center of the marker and the image center was 0.31 mm (range, 0.14-0.49 mm).
No angular dependence of the offsets on the gantry or couch rotations was observed. The same tendency was observed for each preset, as shown in Table 4.

| DISCUSSION
This report is the first of an evaluation of the accuracy of the Sync- <0.2°for the phantom, and <1.5 mm, <1.0°for the patients. 6 They also noted that the impact of rotation on the differences was minor  report 19 and thus a clinically acceptable value. As recommended by the AAPM TG-180 report, 20 further clinical investigations are necessary from the viewpoint of positional accuracy and radiation dose.
The end-to-end test in the present study revealed that the positional accuracy (<1 mm) required for intracranial SRT was Each test in the present study was conducted in accordance with the AAPM TG-142 report. It is important to evaluate both the trends in image quality and the coincidence between the imaging coordinate and that of the treatment systems for monthly quality assurance (QA), and the radiation dose for the annual QA. The TG-142 report does not provide a recommendation about how frequently the degree of coordinate coincidence between two verification systems (such as CBCT and the SyncTraX FX4 system) should be measured, but we recommended that such a measurement should be conducted every 6 months.
In the present study, we focused on intracranial SRT and conducted a clinical commissioning of the SyncTraX FX4 system. However, each of the FPDs of the SyncTraX FX4 system has an effective field of view of 15 × 15 cm at the isocenter, and then can be used to perform positional verification for the chest and pelvic regions.
The commissioning process described herein will be a reference for planning and executing a commissioning at each institution regardless of tumor sites.

| CONCLUSION S
We evaluated the accuracy of the SyncTraX FX4 system through a clinical commissioning for intracranial SRT. The results of our analyses demonstrated that intracranial SRT using this system can be realized with clinically acceptable accuracy.

ACKNOWLEDG MENTS
This research was supported by Japan Society for the Promotion of Science (JSPS) KAKENHI grants, nos. 18K15584 and 18K15626.

CONFLI CT OF INTEREST
The authors have no conflicts of interest to declare.
F I G . 9 . Relationship between the tube current and air kerma in (a) Preset #3 and (b) Preset #4.
F I G . 1 0 . The images from the end-to-end test. The numbers followed by "G" and "C" are the gantry and couch angles respectively.
T A B L E 4 Median value of the offset between the center of the marker and the image center for each preset in the end-to-end test.