3. myQA Machines - CBCT QA Plugin

Introduction

The CBCT (Cone Beam Computed Tomography) is commonly integrated into modern linear accelerators. It is a necessary part of an overall QA program as recommended by AAPM TG-142.

CBCT QA is generally performed by capturing a CT scan through a specially designed phantom. These phantoms have many segregated regions to test different aspects of image quality. Important aspects of image quality include contrast, spatial resolution, uniformity, and noise.

The CBCT QA Plugin is designed to analyze the routine CT / CBCT image quality and give reproducible results. Its calculations are intended to satisfy the routine QA specified in the following standards:

• AAPM TG 142: Quality assurance of medical accelerators[1]
• DIN 61223-2-6: Leistungsmerkmale zur Bildgebung von Röntgeneinrichtungen für die
  Computertomographie[2]
• Quality Control in Cone Beam Computed Tomography EFOMP-ESTRO-IAEA Protocol[3]

CBCT QA Plugin Test Types

The CBCT QA Plugin includes the following test types:

Table 4.1. The CBCT QA Plugin test types and test parameters

The CBCT QA Plugin provides Individual tests for each test type and a combined test for all test result
parameters.

Supported CBCT phantoms

The CBCT QA Plugin supports the following phantoms:

• CatPhan® 503, 504, 600 and 604
• CIRS 610 AAPM
• Gammex 464 and 467
• GE QA phantom
• Tomo Cheese

For all these phantoms, the modules, the ROIs, the materials, and the corresponding HU are predefined. However, the user can modify any of these parameters in the Test Setup page. The user can also define new phantoms.

Workflows for CBCT QA Plugin Tests

Note: For detailed information on the myQA - Imaging QA – TestSetup/TestRun CBCT window, see Section             4.1, myQA Machines Reference Guide.

Setting up CBCT QA Baselines in Test Setup

The test settings and baselines are defined in the Test Setup workspace and cannot be modified in the Test Run workspace.

• In Test Setup, click Edit in the Test Setup ribbon to enter the Edit mode.
• Select the desired machine in the Machines & Templates panel.
• In the Protocol Tree panel, click New in the ribbon to create a task.
• When the task is selected, click New in the ribbon to create tests:

New Test Definition

• In the New Test Definition dialog (see next page), enter the required name of the test and optionally a description.
  New Test definition window
• Select CBCT kV/MV for Category (1) (for information display only)
• Select one of the CBCT Result Type (2), e.g., CBCT: Contrast for an individual test or CBCT: Combined for all test result types.
• Define Energy (3): e.g., select the kV round button and enter the value of the kV.
  Note that when Undefined is selected, the exact value of the energy cannot be entered.
  Note: Only tests with the same defined energy or those with Undefined selected can be shared.
           Migrated tests from a previous version will have an undefined energy by default. A manual                   adjustment in the Test Setup is advised to match the energy type that was used and value.
• Click OK (4). The test is listed in the Protocol Tree panel and displayed in the Test Definition panel.
• Repeat the previous steps to create more tests as needed.

Define the test baseline

• Select a test in the protocol treeTest Setup workspace
• In the Test Definition panel, click Open Module (5). The myQA - Imaging QA – TestSetup CBCT window opens with default settings.
• In the Equipment setup section > Phantom image(s) combo box, import the image or image stack as reference images. The imported image(s) will be displayed in the Image section.
• myQA - Imaging QA – TestSetup CBCT window
• Once the images are imported and the ROIs have been already defined (e.g., a SW supported phantom), the SW automatically selects the correct slice and assigns it to a designated module as a reference image. It is recommended to check whether the auto-slice selection is correct:
  
  • Select a phantom module in the Modules box. In the Images section, check whether the predefined ROIs are correctly placed. Do this for all modules.
  • If the ROIs are only slightly miss-placed, use the functions provided in the Tools panel to move the ROIs.
  • If a different slice needs to be selected, scroll through the slices to search for the right image for this module. Once found and the ROIs are also properly defined, click the Is reference button.
    If the ROIs need to be defined or re-defined, see Section 4.1, myQA Machines Reference Guide for detailed instructions.
• After reference images for all modules are set, click the Process button in the Baseline values section. The Expected and tolerance (Warn and Fail) values will be filled. Edit the tolerances if necessary. They can be edited by changing the values in the corresponding boxes.
• Click the OK button. The reference images and configuration files are listed in the Test Definition
  section.
• Repeat the above “Define the test baseline” steps for CBCT QA test if desired.
• Click Set Active in the ribbon if the protocol is not yet set as an active protocol.
• Click Save in the ribbon.
  Note: For this step, you can also create a test template first, and then copy the template into the
  machine protocol.

Execute CBCT QA Tests in Test Run

• Click the Test Run button in Navigation Panel.
• In the Agenda panel, select the corresponding protocol and task.
• In the Task View panel, select a test.
• Optionally, for an individual test, if the Share Results checkbox is checked the results will be shared by other CBCT tests in the test list. When this test is finished, other CBCT tests that use the same image will also be finished automatically. The Share Results checkbox is irrelevant for a combined test.Test Run workspace (before tests are executed)
• Click the Start button. The myQA – Imaging QA – TestRun CBCT window opens.
  Hint: If no setup is saved, the Start button in Test Run is disabled.
• In the Equipment setup section > Phantom image(s) combo box, import the image or image stack
  as reference images. The software loads all images taken for each module and displays in the Image
  section.
  As importing a reference image, once the images are imported, the software automatically selects the correct slice based on the ROIs and assigns it to a designated module as a test image (active image). It is recommended to check whether the auto-slice selection is correct:
  
  • Select a phantom module in the Modules box. In the Images section, check whether the image is
    similar to the reference image and the ROIs are correctly placed. Do this for all modules.
  • If the ROIs are only slightly miss-placed, use the functions provided in the Tools panel to move the ROIs.
  • If a different slice needs to be selected, scroll through the slices to search for the right image for this module. Once found and the ROIs are also properly defined, click the Is active image button.
•  If necessary, the Image view can be displayed in a full window by clicking the icon and the user can zoom in on the image.
•  Once the test images for all modules are properly selected, click Process in the Baseline values
  section. See the description of test results in Section 4.1.5, myQA Machines Reference Guide.myQA – Imaging QA – TestRun CBCT window
• Click OK; results for executed CBCT tests that share the same module are displayed if the Share Results checkbox is checked.Test Run workspace (after tests are executed)
• Repeat the above step to execute the next test until all tests are finished.
• Click Ok to close the myQA – Imagine QA – TestRun CBCT window.
• Click Finish in the Test Run ribbon and then finish in the Finish Task dialog.
  Afterwards, the test results will appear in myQA Cockpit and a report can be created in the Test Repository page.Test Repository page with executed test list

References

[1] F. Klein, "Task Group 142 report: Quality assurance of medical accelerators," Med. Phys., vol. 36, p. 4197, 2009.

[2] DIN, "Leistungsmerkmale zur Bildgebung von Röntgeneinrichtungen für die Computertomographie," DIN EN 61223-2-6, 2008.

[3] H. de las Heras-Gala, "Quality Control in Cone Beam Computed Tomography EFOMP-ESTRO-IAEA
Protocol," http://dx.medra.org/10.19285/CBCTEFOMP.V1.0.2017.06, 2017.

[4] ELEKTA, "XVI 5.0 Customer Acceptance Tests," 2013.

[5] Lehmann, "Commissioning experience with cone-beam computed tomography for image-guided radiation therapy," Journal of Applied Clinical Medical Physics, p. 21, 2007.

[6] Kaur, "Image Recognition using Coefficient of Correlation and Structural Similarity Index in Uncontrolled Environment," International Journal of Computer Applications, vol. 59, no. 5, p. 32, 2012.