Room: Exhibit Hall | Forum 6
Purpose: Quality assurance of multileaf collimator (MLC) for modern radiation therapy require sub-millimeter resolution. Traditional methods to perform MLC QA rely on: (1) specified MLC type, (2) an extra alignment tool for assisting leaves identification, (3) fixed relative orientations and positions among the calibration tool, gantry, and collimator, and (4) manual operations, which complicates the QA procedure; limits testing items; and brings in randomized artificial error. To circumvent the adversity, an automate QA system, named visualized QA (vQA), is proposed for general MLC QA.
Methods: A radioluminescent phantom covered with a mixture of Gd2O2S:Tb and PDMS, which enables visualization of the radiation beam on the same surface, is readily detachably mounted on the gantry. A customized treatment plan that controls LINAC was written to implement the delivery sequences for a modified picket-fence test. The emitted light was subsequently captured using a time-gated camera, and images were processed to extract leavesâ€™ size and position information.
Results: Tests were performed for the HD120 MLC and the Millennium MLC (Varian Medical Systems, Palo Alto, CA). A resolution of 0.3mm can be obtained for the MLC projection image, and all the testing results match well (a maximum disagreement of 0.1 mm) with those via a commercial MLC QA instrument (Standard Imaging, Inc.). Total measurement time was less than 10 minutes for all tests.
Conclusion: This work demonstrates a "one-click" automating MLC QA system, whose unique features of (1) complete QA tests for different combinations of gantry position and collimator orientation; (2) no connection with the LINAC for synchronization; and (3) promising broad compatibility for multiple MLC types via leaf automatic identification technique.