Room: AAPM ePoster Library
We created a computer graphical user interface (GUI) for the dosimetrist and medical physicist to manipulate the shape between adjacent multileaf collimator (MLC) apertures using a novel leaf-sequencing approach. Moreover, we integrated the GUI into a treatment planning system (TPS).
The GUI named “ShapeX” was written in Visual C++. In the “ShapeX” GUI, the accuracy of algorithm to generate the MLC apertures was justified by the deviation between the intensity distributions of the photon field calculated and the distributions obtained from the optimization system in the TPS. In our algorithm, we developed a shape optimization technique that the MLC aperture is subjected to the distances of the adjacent rows, and is controlled by a shape constraint d. To verify and test our algorithm, a set of benchmark calculations were performed.
Intensity fluence maps generated by our “ShapeX” GUI were compared to the related code system based on the leaf-sequencing algorithm from Xia and Verhev. Results from the benchmark of intensity maps showed that our algorithm performed better with smaller number of beam segments, monitor unit per segment and leave travel distance. When “ShapeX” was integrated into the KylinRay-IMRT TPS. At a button-click, MLC segments can be generated and displayed almost instantly. By changing the shape constraint in the GUI, user can explore its effects on the resulting MLC aperture shapes.
A GUI powered by a leaf-sequencing algorithm using a novel shape optimization technique was validated with a series of intensity maps. The algorithm was integrated into the KylinRay-IMRT TPS. Our results show that the improved algorithm has good agreement with the benchmark and is faster than the classical method. Moreover, the algorithm can keep the MLC aperture in a compact shape.
Funding Support, Disclosures, and Conflict of Interest: Teaching research project of education bureau of Anhui province, China (2018jyxm1001)
MLC, Intensity Modulation, Computer Software
TH- External Beam- Photons: IMRT/VMAT dose optimization algorithms