An evaluation of the stability of image quality parameters of Elekta X‐ray volume imager and iViewGT imaging systems

Abstract Introduction A robust image quality assurance and analysis methodology for image‐guided localization systems is crucial to ensure the accurate localization and visualization of target tumors. In this study, the long‐term stability of selected image parameters was assessed and evaluated for the cone‐beam computed tomography (CBCT) mode, planar radiographic kV mode, and the radiographic MV mode of an Elekta VersaHD. Materials and Methods The CATPHAN, QckV‐1, and QC‐3 phantoms were used to evaluate the image quality parameters. The planar radiographic images were analyzed in PIPSpro™ with spatial resolution (f30, f40, f50), contrast to noise ratio (CNR) and noise being recorded. For XVI CBCT, Head and Neck Small20 (S20) and Pelvis Medium20 (M20) standard acquisition modes were evaluated for uniformity, noise, spatial resolution, and HU constancy. Dose and kVp for the XVI were recorded using the Unfors RaySafe Xi system with the R/F low detector for the kV planar radiographic mode. For each metric, values were normalized to the mean and the standard deviations were recorded. Results A total of 30 measurements were performed on a single Elekta VersaHD linear accelerator over an 18‐month period without significant adjustment or recalibration to the XVI or iViewGT systems during the evaluated time frame. For the planar radiographic spatial resolution, the normalized standard deviation values of the f30, f40, and f50 were 0.004, 0.003, and 0.003 and 0.015, 0.009, and 0.017 for kV and MV, respectively. The average recorded dose for kV was 67.96 μGy. The standard deviations of the evaluated metrics for the S20 acquisition were 0.083(f30), 0.058(f40), 0.056(f50), 0.021(Water/poly‐HU constancy), 0.029(uniformity) and 0.028(noise). The standard deviations for the M20 acquisition were 0.093(f30), 0.043(f40), 0.037(f50), 0.016(Water/poly‐HU constancy), 0.010(uniformity) and 0.011(Noise). Conclusion A study was performed to assess the stability of the basic image quality parameters recommended by TG‐142 for the Elekta XVI and iViewGT imaging systems. The two systems show consistent imaging and dosimetric properties over the evaluated time frame.

a small FOV will be used with a 200°rotation (comparable to a fullfan CBCT) for imaging of the head or neck while the medium FOV (Half-fan CBCT) is standardly used for larger sites. When a medium or large FOV is selected, the detector panel is shifted 11.5 cm and 19 cm, respectively, from the central axis of the kV X-ray beam (the small FOV is obtained by centering the detector pane). 8 The XVI contains preset parameters that are configured per anatomical site for imaging geometry, beam characteristics, and reconstruction method. It also allows for customization of the tube potential, number of frames, mA and ms per frame, start and stop gantry angles, and reconstruction resolution (1-mm pixel size for medium resolution and 0.5-mm pixel size for high resolution). For this portion of the study, a 200°gantry rotation with small FOV will be analyzed along with a 360°gantry rotation with a medium FOV.

2.A.2 | Elekta iViewGT electronic portal imaging device
The Elekta iViewGT (Elekta, Crawley, UK) is an amorphous silicon flat panel imaging device mounted on a robotic arm designated at C in Fig. 1. This arm allows the detector to be positioned at source to electronic portal imaging device (EPID) distance of 160 cm with an active imaging area of 41 9 41 cm. 9 The image matrix is created from an array of 1024 9 1024 photodiodes with a pitch of F I G . 1. The X-ray volume imaging (XVI) guidance system and iViewGT image system of the Elekta VersaHD radiation delivery system are shown. (a) a-Si flat panel detector of the XVI, (b) kV Xray source of the XVI and (c) iViewGT imaging panel.
modes, a single exposure, 6 MV planar radiographic mode was used in this study.

2.A.3 | The CATPHAN â 504 Phantom
The CATPHAN 504 (Phantom Laboratory, Salem, NY, USA) was used to evaluate the image quality parameters of the kV-CBCT for both small and medium acquisition modes. The CATPHAN is a cylindrical phantom with outer diameter of 20 cm, inner diameter of 15 cm and 4 different inserted modules that can evaluate image uniformity, image noise, image high contrast spatial resolution, HU constancy, geometric distortion, and slice thickness. 11 The CATPHAN was chosen for its ease of setup and use, commercial availability (commonly provided with purchase of linear accelerator) and compatibility with the PIPSpro software. and an image of the QCkV-1 or QC-3 phantoms or (2) acquire two sequential phantom images. In this study, the images were evaluated using an acquired flood field and one image of the phantom.

2.A.5 | Unfors RaySafe Xi R/F and CT Detectors
The Unfors RaySafe Xi (Unfors RaySafe AB, Billdal, Sweden) is a comprehensive system of detectors that can perform multi-parameter measurements on all X-ray modalities. The system is composed of a base unit and multiple detectors that are jointly certified by the AALA (American Association for Laboratory Accreditation) and ADCL (American dosimetry calibration laboratory). In this study, the R/F was used in conjunction with the base unit for kV planar radiographic mode. The R/F detector is a small, lightweight, portable, and wireless detector capable of measuring kVp, dose, dose rate, pulse, pulse rate, dose/frame, time, half value layer, total filtration and waveforms simultaneously. For the purposes of this study, the image parameters evaluated were the dose and the X-ray energy for the kV planar radiographic mode.

2.B.1 | kV planar radiographic
To evaluate the imaging quality parameters, the QCkV-1 phantom was placed directly onto the face of the XVI detector with the F0/ S20 inserts and aligned to the room lasers as seen in Fig. 2. One image was acquired with the following settings: 70 kV, 160 mA, and 200 ms. After removing the QCkV-1 phantom, a second flood field image was acquired with the same settings as before. The two images were then analyzed in PIPSpro TM and the high contrast spatial resolution, noise and contrast to noise ratio were recorded. Each image has three separate values of the high contrast spatial resolution (f30, f40, f50(lp/mm)), which represent the frequencies at 30%, 40% and 50% of the maximum for the relative modulation transfer function (RMTF).
Next, the Unfors RaySafe Xi R/F detector was placed onto the XVI detector. The process was repeated with the dose and X-ray energy being manually recorded after each acquisition.

2.B.2 | MV planar radiographic
To evaluate the imaging quality parameters, the QC-3 phantom was
T A B L E 5 Image quality consistency thresholds for the Medium (M20) CBCT.
in this study. More investigation into the effect of these temporal differences and threshold limits is needed to quantify the effect an out of tolerance measurement would have clinically. As the image quality and technology of CBCT continues to improve, the clinical impact of the temporal image quality deviations needs further evaluation. In general, technological advances, including advances in detector design, generator output consistency or image reconstruction algorithm, will require careful consideration on a case by case basis as to the effect on the clinically established baselines.

| CONCLUSION
A study of the stability for image quality parameters of Elekta XVI and iViewGT imaging systems was performed using commercially available imaging QA phantoms and software with a total of 30 measurements over an 18-month period. Run charts were created for each of the evaluated parameters. Both systems, for each image quality parameter, show consistent imaging and dosimetric properties over the evaluated time frame for the normalized mean and standard deviations, as well as comparable results to previously completed studies. 7-9

CONFLI CT OF INTEREST
The authors have no conflict of interest to report.