Commissioning of the Mobius3D independent dose verification system for TomoTherapy

Abstract In radiation therapy, a secondary independent dose verification is an important component of a quality control system. Mobius3D calculates three‐dimensional (3D) patient dose using reference beam data and a collapsed cone convolution algorithm and analyzes dose‐volume histogram automatically. There are currently no published data on commissioning and determining tolerance levels of Mobius3D for TomoTherapy. To verify the calculation accuracy and adjust the parameters of this system, we compared the measured dose using an ion chamber and film in a phantom with the dose calculated using Mobius3D for nine helical intensity‐modulated radiation therapy plans, each with three nominal field widths. We also compared 126 treatment plans used in our institution to treat prostate, head‐and‐neck, and esophagus tumors based on dose calculations by treatment planning system for given dose indices and 3D gamma passing rates with those produced by Mobius3D. On the basis of these results, we showed that the action and tolerance levels at the average dose for the planning target volume (PTV) at each treatment site are at μ ± 2σ and μ ± 3σ, respectively. After adjusting parameters, the dose difference ratio on average was −0.2 ± 0.6% using ion chamber and gamma passing rate with the criteria of 3% and 3 mm on average was 98.8 ± 1.4% using film. We also established action and tolerance levels for the PTV at the prostate, head‐and‐neck, esophagus, and for the organ at risk at all treatment sites. Mobius3D calculations thus provide an accurate secondary dose verification system that can be commissioned easily and immediately after installation. Before clinical use, the Mobius3D system needs to be commissioned using the treatment plans for patients treated in each institution to determine the calculational accuracy and establish tolerances for each treatment site and dose index.

(3D) dose distribution for a patient using computed tomography (CT) datasets that employ information from the RT plan after receiving the DICOM CT datasets, RT Plan, RT Structure, and RT dose from the TPS. Following those calculations, Mobius3D automatically compares the dose computed by the TPS with that calculated by Mobius3D. Finally, Mobius3D indicates pass/fail results for the dose-volume histogram (DVH) limits and the 3D gamma passing rate. 12,13 The user can check the results in a variety of ways-such as the DVH, dose index, 3D dose distribution, dose profile, and gamma distributionfrom any place in the network using a web browser.
Mobius3D has unique characteristics that utilize the reference beam data. The system does not require specific measurement data for TomoTherapy, and it can therefore be installed immediately.
Another characteristic is that it uses a collapsed cone convolution superposition algorithm 14,15 developed by the manufacturer that is accelerated through graphics processing units 16,17 for the dose calculation. This algorithm can produce accurate calculations for IMRT and for heterogeneous conditions. 18 These two characteristics make it possible for a user to validate the TPS dose calculation and provide 3D dose distributions and DVH information immediately and independently from the TPS.
The accuracy of the dose calculation from a verification system must be commissioned and validated before clinical use.
Commissioning and clinical implementation for Mobius3D has been reported for C-arm linear-accelerator photon beams, [19][20][21] but this has not previously been done for TomoTherapy.
Mobius3D includes parameters for modifying the initial registered data if the machine data in the user institution do not correspond to the reference beam data. The purpose of the present work is to describe our experience with these adjustment parameters and to evaluate the accuracy of dose calculations for TomoTherapy. We also indicate the action and tolerance levels for clinical implementation.

2.A | Mobius3D system for TomoTherapy
The TomoTherapy system has three nominal field widths (NFWs) in the superior-inferior direction of the patient: 1.0, 2.5, and 5.0 cm jaw size at the source-to-axis distance 85 cm. To adjust the reference beam data, Mobius3D has three parameters for TomoTherapy: • the calibration value, • the output factor for NFW 1.0 cm × 40 cm normalized to NFW 5.0 cm × 40 cm, and • the output factor for NFW 2.5 cm × 40 cm normalized to NFW 5.0 cm × 40 cm.
The output factor for NFW 5.0 cm × 40 cm is locked at 1.0, and the user cannot change this in Mobius3D. If this output factor needs to be changed, the Mobius support team must provide a new license file that has an arbitrary calibration value. This file affects all the NFW conditions, so the user must recalculate and validate all plans.
Thus, to adjust the parameters for all NFW conditions, it is first necessary to prepare the license file adjusted to NFW 5.0 cm and then adjust the two output factors to NFW 1.0 and 2.5 cm.

2.B | Dosimetric verification
We used a water equivalent phantom called "cheese phantom" that was provided by Accuray for our TomoTherapy. This phantom is a cylinder 30 cm in diameter and 18 cm long. It can contain a cylindrical ion chamber that can be positioned at an arbitrary location along the perpendicular axis to the cylinder axis and a film in half of the phantom. To evaluate the dose calculation accuracy for patient plans, we imposed the contours of patients who have been treated at our institution on this phantom. In the present study, we evaluated nine helical IMRT plans for each NFW. These plans include three each for prostate, head-and-neck, and pelvic lymph node cases.
We calculated these plans using Planning Station version 5.1.0.4 on the basis of our clinical dose constraints, and we sent the DICOM datasets to Mobius3D to calculate the verification dose. We used an ion chamber with an active volume of 0.057 cc (Exradin A1SL ionization chamber; Standard Imaging Inc., Middleton, WI, USA) to measure two points in the PTV for comparison with the doses calculated by Planning Station and by Mobius3D. We contoured the A1SL sensitivity volume in the PTV and recorded the mean dose delivered to that volume. These volumes were placed at a low gradient region such that the ratio of maximum dose to minimum dose in the sphere of radius 7.0 mm centered at the sensitivity volume was over 0.96. 22 We defined the dose difference ratio between measurement and calculation as follows: where D calc is the dose as calculated by Mobius3D or Planning Station, and D meas is the dose as measured using the ion chamber. To measure the relative dose distribution, we used GAFCHROMIC EBT3 Film (Ashland ISP Advanced Materials, NJ, USA). A film was placed in the sagittal plane to measure the planar dose distribution.
We extracted the calculated dose distribution from the DICOM RT dose using an in-house program, and we performed gamma analysis using the RIT 113 film dosimetry system (Radiological Imaging Technology, Inc., Denver, Colorado, CO, USA). Each dose distribution was normalized to the mean dose of the area in the PTV. We performed gamma analysis using the criteria of 3% and 3 mm, which are the dose difference relative to the global maximum dose and the distance-to-agreement values, respectively, using a 10% threshold to exclude the low-dose region. We calculated the four indices mean (μ), standard deviation (σ), maximum (Max), and minimum (Min) for each dose difference ratio and gamma passing rate.

2.C | Clinical implementation
To evaluate the differences between Mobius3D and Planning Station in a clinical situation, we recalculated the treatment plans using Mobius3D for patients who had been treated in our institution.
This study was approved by the institutional review board in our institution (number 836). We included the following numbers of plans and treatment sites, with the corresponding NFWs: 29 prostate plans with NFW 1.0 cm, 27 prostate plans with NFW 2.5 cm, 62 head-and-neck plans with NFW 2.5 cm, and 8 esophagus plans with NFW 2.5 cm. Prescription protocol was that 95% of the PTV was received as the prescription dose. Mobius3D recalculated the 3D dose distribution without renormalization. We obtained the following dose indices for the PTV (D mean , D 2% , D 50% , D 95% , and D 98% ), rectum (D mean , V 65Gy , and V 40Gy ), bladder (D mean , V 65Gy , and V 40Gy ), spinal cord (D max ), each parotid (D mean ), and each lung (D mean , V 20Gy , and V 5Gy ). We defined the dose difference ratios and volume differences for each dose index between Planning Station and Mobius3D as follows: where D M3D is the dose as calculated by Mobius3D, D PS is the cor-

2.D | Action and tolerance level
The action and tolerance levels for the dose differences between Planning Station and Mobius3D were set at μ ± 2σ and μ ± 3σ, respectively. We calculated the mean dose for the PTV at each treatment site and the mean dose for the OAR at all treatment sites.

3.A | Dosimetric verification
In this report, we optimized three parameters on the basis of verification using the ion chamber for each NFW. We adjusted the calibration value, which we increased to 3% from the initial registered data, and we increased the output factor for NFW 1.0 cm by 2.5% from initial registered data. to −2.7. The standard deviation is similar to Mobius3D. Table 2 shows the gamma passing rates using the criteria of 3%    3.C | Action and tolerance level Figure 2 shows a histogram of the mean dose discrepancies for the PTV and the OAR between Mobius3D and Planning Station. The standard deviation was similar, however, the median is different from the head-and-neck plans in Fig. 2(a). Table 5 shows the action and tolerance levels, which was set at μ ± 2σ and μ ± 3σ, respectively, for the mean dose discrepancies between Mobius3D and Planning Station.

| DISCUSSION
Using reference beam data has the benefits that the verification system is independent from the TPS and that is can be implemented immediately in the clinic. It is important that beam data of linear accelerator in the user institution be consistent with the reference beam data in Mobius3D. In this work, we used the method written T A B L E 1 Summary of differences between measurements using ion chamber and calculations by two systems.  [25][26][27][28] Differences in these two implementations were more apparent in pelvic lymph cases. Source size affects penumbra shape and MLC leakage affects organ at risk. 29,30 We have also determined the differences between Mobius3D and Planning Station for the DVH, dose indices, and 3D gamma ing accuracy and beam data is indicated in the phantom study. In Table 1, Mobius3D values were larger than Planning Station at point dose in PTV. In addition, calculation accuracy in heterogeneity condition affects these results in the patient study. Figure 3   Rectum D mean 0.5 ± 0.6 (1.8/−0.2) 0.6 ± 0.9  There is no guideline for setting the action and tolerance levels for a 3D secondary independent verification system. Instead, it is  verification. This indicates thatto set the action and tolerance levels for 3D independent verificationthe user must analyze data for each treatment site and dose index. Our results are similar to previous results that analyzed differences between a TPS and Mobius3D for C-arm linear accelerators for IMRT and non-IMRT. 21 All tolerance levels were within 5% total uncertainty, which is a widely accepted goal for effective radiation treatment. 32,33 Patient-specific quality assurance needs systems to verify all treatment processes so as to accurately detect an error that could harm the patient. 34 In addition to treatment planning process verification using Mobius3D, data transfer from TPS to linear accelerator control system and hardware malfunction during irradiation must be verified. In recent research, new approach was opposed that the system verifies the dose distribution between TPS and reconstructed dose from information generated from linear accelerator after irradiation. 24,[35][36][37] In addition to Mobius3D, Mobius has a system using these new approaches named MobiusFX for C-arm linear accelerators, but it does not deal with TomoTherapy now. 38,39 To secure the patient safety in TomoTherapy, we should use systems implementing these new approaches or conventional measurementbased techniques in addition to a secondary independent dose verification system.

| CONCLUSION
We described our experience with commissioning Mobius3D for TomoTherapy, and it revealed that Mobius3D has enough accuracy for the independent dose calculation system in our institution. As a result of our experience, we conclude that Mobius3D can be commissioned easily and immediately after being installed. We also indicated the action and tolerance levels for clinical implementation. A 3D secondary independent verification system gives us much more useful information about each target and OAR than we have possessed before. However, it is difficult to decide whether each plan is acceptable or not without commissioning and setting tolerances.
Before clinical use, each clinic should commission the system to determine the calculation accuracy and to establish tolerances for each treatment site and dose index using treatment plans for patients who have been treated in that institution.