Investigation of lnter and Intra-fractional Uncertainties in Lung IMRT Delivery

Maria Das KJ; Arpita Agarwal; Yoganathan SA; Gowtharu Raj D; Shaleen Kumar

doi:10.3969/j.issn.1004-0552.2017.01.001

摘要：Changes in the tumour position due to inter and intra-fractional motion introduce uncertainty in IMRT delivery. In this work, we experimentally evaluated the effect of inter and intra-fractional motion uncertainties in the IMRT delivery. A lung patient was planned with rive field dynamic IMRT for a total dose of 60 Gy in 30 fractions using 6 MV photon beam at a dose rate of 400 MU/min. The plan was delivered on varian CL2100CD linear accelerator with millennium 120 MLC for a single fraction and measured using IMRT MatriXX placed on Quasar motion platform, aligned with respect to isocentre for planar as well as point dose measurements. This measurement was summated for 30 fractions and taken as a reference. 30 measurements were performed for each inter-fractional, intra-fractional and combined effect. The setup errors (Mean+SD) used to simulate the inter-fractional displacements were RL:-0.10+0.27 mm, SI: 0.17+0.45 mm, AP: -0.04+0.38 mm and Rot: 0.02+0.86 degree. The intra-fractional motion was simulated using the motion platform parallel to the MLC leaf motion, for an amplitude of 1 cm and a period of 4 s. The planner fluence of inter, intra-fractional motion and combination of both was analyzed against the reference using gamma criteria of 3％/3 mm. Similarly, the point dose measurements were also compared. The maximum deviation in point dose during a single fraction was-3.8％, 3.1％and-5.9％for inter-fractional, intra-fractional and combined respectively. The same resulted in deviation of-0.4％,-0.3％and 0.1％respectively when summated for 30 fractions. The percentage of pixels failing the gamma criteria during a single fraction was 28.0％, 16.4％and 28.5％for inter-fractional, intra-fractional and combined respectively. The same revealed 11.4％, 12.4％and 19.2％respectively when summated. Though the point dose deviations were nullified over 30 fractions, the planner fluence variation was observed to be considerable.