From one treatment to the next, considerable effort is made to accurately position radiotherapy patients according to their treatment plans. However, some variation is unavoidable. The target volume and the organs at risk may also move within the patient and/or change shape during the treatment. Thus, it is important to be able to verify the success of the treatment by determining the position of patient and the dose deposited in the patient at each fraction. One possibility to achieve this, particularly when equipment, time and budgets are limited, would be to collect limited information while the patient is on the treatment couch. This research was aimed to develop a method for optimum beam position determination, for each patient-specific case. The optimized beam positions would balance the both treatment planning and image reconstruction, so that the patient???s image can be obtained during the treatment delivery using the information collected from the same angles as used for treatment. This will allow verification of the dose deposited in the patient for every fraction. Using a limited number of angles for image reconstruction, the dose to the patient can be minimized. This work has two major parts, beam position optimization for image reconstruction and beam position optimization for treatment planning. These two optimizations are then combined to obtain the optimum beam position for both image reconstruction and treatment planning. An objective function, projection correlation, was developed to investigate the image reconstruction method using limited information. Another objective function, the average optimization quality factor, was also introduced to optimize beam positions for treatment planning. Two optimization methods, the gradient descent method and the simulated annealing based on these objective functions were used to determine the beam angles. The results show that the projection correlation presents several advantages. It can be applied without any iterations, and it produces a fast algorithm. The present research will allow selection of the optimum beam positions without excessive computational cost for treatment planning and imaging. By combining the projection correlation and the average optimization quality factor together with more advanced image reconstruction software this could potentially be used in a clinical environment.
Identifer | oai:union.ndltd.org:ADTP/187139 |
Date | January 2006 |
Creators | Widita, Rena, Physics, Faculty of Science, UNSW |
Publisher | Awarded by:University of New South Wales. School of Physics |
Source Sets | Australiasian Digital Theses Program |
Language | English |
Detected Language | English |
Rights | Copyright Rena Widita, http://unsworks.unsw.edu.au/copyright |
Page generated in 0.0016 seconds