AbstractID: 8181 Title: An algorithm to incorporate leaf transmission and head scatter corrections into step-and-shoot leaf sequences for IMRT

ID: 8181 Title: An algorithm to incorporate leaf transmission and head scatter corrections into step-and-shoot leaf sequences for IMRT Leaf transmission and head scatter are two most important factors that influence IMRT delivery and should be fully taken into account when generating MLC leaf sequences. Significant discrepancies between the desired and delivered intensity profiles could otherwise be resulted. This work presents a reliable algorithm to minimize the inaccuracy caused by the two factors in step-and-shoot mode. In this algorithm, a three-source model is used to calculate the relative head scatter distributions for each segment, and a downhill simplex optimization method is employed to minimize the difference between the desired and computed fluence maps (in a leastsquare sense) by adjusting the fractional MU factors in the leaf sequence file generated using conventional method without considering the two effects. In general, head scatter correction leads to a reduction of incident intensity and thus increase of beam-ontime, especially for complicated intensity maps with a large number of small segments; conversely, the transmission correction increases the incident intensity and requires reducing beam-on-time. In some circumstances, due to the opposite effects of the two factors, larger errors may be resulted if only one factor is considered than if both effects are ignored. The algorithm was tested by experimental measurements using several clinical IMRT cases. The experimental data show that the dose distributions delivered by the leaf sequences optimized by our algorithm agree with the desired ones within 2.5% or 2mm in high dose gradient regions, significantly improving the leaf sequences provided by commercial planning systems and providing a reliable tool for clinical IMRT application.