Poster - Thurs Eve-23: Effect of lung density and geometry variation on inhomogeneity correction algorithms: A Monte...

This study provides new information on the evaluation of the lung dose calculation algorithms as a function of the relative electron density of lung, ρ_e,lung . Doses calculated using the collapsed cone convolution (CCC) and adaptive convolution (AC) algorithm in lung with the Pinnacle³ system were compared to those calculated using the Monte Carlo (MC) simulation (EGSnrc-based code). Three groups of lung phantoms, namely, "Slab", "Column" and "Cube" with different ρ_e,lung (0.05-0.7), positions, volumes and shapes of lung in water were used. 6 and 18MV photon beams with 4×4 and 10×10cm² field sizes produced by a Varian 21EX Linac were used in the MC dose calculations. Results show that the CCC algorithm agrees well with AC to within ±1% for doses calculated in the lung phantoms, indicating that the AC, with 3-4 times less computing time required than CCC, is a good substitute for the CCC method. Comparing the CCC and AC with MC, dose deviations are found when ρ_e,lung are ⩽0.1-0.3. The degree of deviation depends on the photon beam energy and field size, and is relatively large when high-energy photon beams with small field are used. For the penumbra widths (20%-80%), the CCC and AC agree well with MC for the "Slab" and "Cube" phantoms with the lung volumes at the central beam axis (CAX). However, deviations >2mm occur in the "Column" phantoms, with two lung volumes separated by a water column along the CAX, using the 18MV (4×4cm² ) photon beams with ρ_e,lung ⩽0.1.