University of Wollongong
Browse

The characterisation of MP512 for 2D/3D dose reconstruction for small beams

Download (20.38 MB)
thesis
posted on 2024-11-12, 13:37 authored by Nauljun Stansook
Stereotactic radiation therapy such as SRS and SBRT utilise multiple beams delivery with a small radiation field and high dose gradients. A quality assurance tool with high stability and linearity of radiation response that can map a 2D dose read out in real time and with a high spatial resolution is needed to accurately verify two-dimensional (2D) pre-treatment dose distributions. This thesis describes two 2D monolithic diode arrays based on different silicon substrates called MagicPlate-512 (MP512) designed and developed by the Centre for Medical Radiation Physics to verify small field dosimetry. The first substrate is based on bulk p-type silicon known as MP512-Bulk and the other is based on the high resistivity of a thin epitaxial layer known as MP512-EPI. MP512 allows real time 2D dose mapping with a high spatial and temporal resolution. Both detector arrays consist of 512 0.5x0.5 mm2 active pixels with a 2 mm pitch that covers an area of 52x52 mm2. The angular response of MP512-Bulk as well as its correction factor were investigated for various field sizes and photon energies. It showed a maximum variation of relative angular response normalised to an incidence beam angle zero at a beam angle of 90° of approximately 18.5±0.5% and 15.5±0.5% for 6 MV and 10 MV photon beams. These results indicate that the angular response is sensitive to the energy, whereas the variation of angular response is less affected by field size. The packaging and intrinsic asymmetry of the monolithic silicon detector array structure are the major elements that affect the angular dependence of an MP512-Bulk.

History

Year

2018

Thesis type

  • Doctoral thesis

Faculty/School

School of Engineering Physics

Language

English

Disclaimer

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.

Usage metrics

    Categories

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC