An experiment validation of treatment plannning system dose calculation accuracy in heterogeneous lung tumour

• Main Author: Lee, Wen Han
• Other Authors: -
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2022
• Subjects: Science::Physics::Nuclear and particle physics
• Online Access: https://hdl.handle.net/10356/156954
• Institution: Nanyang Technological University

In the recent years, improvements in imaging technology have allowed for the detection of small tumours in patients. These small tumours that can be found in areas such as the lung can be treated with Stereotactic Body Radiotherapy (SBRT), which requires high accuracy in dose delivery with the high doses used in the treatment. In this study, the accuracy of Anisotropic Analytical Algorithm (AAA) and Acuros XB (AXB) on small tumour targets in a heterogeneous lung phantom was explored. Water-equivalent tumour targets of 0.5 cm, 1 cm, 1.5 cm, 2 cm, and 3 cm were irradiated with square field sizes from 5 cm by 5 cm down to square field sizes with width that is the nearest integer value larger than the diameter of the target, using 10FFF, 10X, 6FFF and 6X photon beams. The same parameters were used to calculate doses in Varian Eclipse TPS (Eclipse Treatment Planning ver13.6, Varian Medical System, Inc, Ca, USA) using AAA and Acuros XB with grid sizes 0.25 cm and 0.1 cm. Doses were also predicted in the TPS with deviations of 3 mm in each axis to account for positional uncertainty during the alignment of the phantom. The results imply that field sizes alone do not greatly affect the accuracy of TPS algorithms. However, positional uncertainty plays a significant role in the accuracy of TPS, accounting up to 11.44% error from a 3 mm position deviation. As the field size decreases bear to the size of the tumour, the effects of positional uncertainty on dose accuracy of TPS becomes much more prominent. The tumour sizes also directly affect the accuracy of the TPS algorithms, where a smaller tumour size will increase the dose inaccuracy. This was deduced to be due to the robustness of dose calculation by the algorithms limited by the grid size used. FFF beams are recommended for SBRT as the dose predictions using FFF beams show a lower percentage error compared to the dose predictions done by using flattened beams.