Year

2010

Degree Name

Master of Science

Department

University of Wollongong. School of Engineering Physics

Abstract

High Dose Rate Brachytherapy is a common modality used in the treatment of intermediate to high risk prostate cancers. Currently, no real-time quality assurance technique is widely used to accurately verify the doses delivered during treatment. In many centres more than one treatment is delivered over a 24 hour time interval with the implant in place. During the time between these fractions the prostate may swell due to the trauma from the needles and the needles may move relative to the prostate.

The main aims of this project are: to determine the characteristics of a commercially available semiconductor diode array and assess whether dose verification is possible on a routine basis; to assess whether prostate oedema significantly effects dose volume histogram statistics for subsequent fractions when the plan generated at the time of theinitial procedure is used for subsequent fractions; and to determine whether oedema is the dominant cause of catheter movements required to reposition the implant with respect to the prostate for subsequent fractions.

The diodes’ dosimetric properties studied included linearity, reproducibility, directional dependence, dose rate dependence and temperature dependence. Linearity and reproducibility were compared to ion chamber measurements while the other characteristics were compared with published data.

The dosimetric study involved 28 patients who were treated at the Mater Hospital in Crows Nest during 2007 and 2008. The patients were all treated in the lithotomy position incorporating a plan developed on a real time planning system based upon ultrasound imaging. As they were treated in this position, a replica of the ultrasound probe, utilised for implantation and imaging, was developed in which the diode array could be placed for the duration of treatment. The position of the diodes inside the replica probe relative to the implant was determined using fluoroscopy so that dosimetric comparisons between the diodes and corresponding positions in the planning system could be made. Ultrasound images taken before each treatment fraction were used to determine the extent of oedema during the course of treatment.

Based upon the measurements examining the diode characteristics and physical set up uncertainty, the overall uncertainty in the patient dosimetric results was determined to be ±10%, which is comparable to published data. Of the patient data points compared,72% showed agreement within this uncertainty. If this was extended to ±20%, assuming measurement uncertainty of ±10% and dose discrepancy of ±10%, then over 95% of the diode results exhibited agreement with the calculated data. A study of the 15 patients studied volumetrically indicated that the mobility of the prostate was patient specific and not dependent upon oedema present, and that the changes in volume that occur during the course of treatment do not significantly effect patient outcomes if the treatment used for the initial fraction is applied to subsequent fractions.

An in-vivo dosimetry program employing a commercial semi conductor diode array could be used to detect gross errors in the delivery of high dose rate prostate brachytherapy as long as the limitations of the device are well understood and the facility has the resources to incorporate such a program.

Download

Share

COinS
 

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.