Modelling geometric uncertainties in prostate radiotherapy

Sage, J 2011, Modelling geometric uncertainties in prostate radiotherapy , PhD thesis, University of Salford.

[img] PDF
Restricted to Repository staff only until 31 July 2022.

Download (35MB) | Request a copy


Image guided radiotherapy (IGRT) presents a range of techniques to enhance the geometric accuracy of radiotherapy treatment. A full evaluation of such techniques, against increased cost, workload, or unnecessary radiation dose, would require the incorporation of geometric errors into existing methods for modelling the effectiveness of radiotherapy treatment. In this work the impact of geometric errors on the tumour control probability (TCP) for prostate radiotherapy is investigated. Positional measurement data are analysed in a manner which differentiates between genuine measured variation and measurement uncertainty. A deformable model of the prostate is developed which encompasses population variation, interfractional variation and observer variability. Ultimately a simulation tool is produced to test different schemes for prostate radiotherapy in the presence of different levels of geometric uncertainty. Measurement error is not generally considered in studies of geometric variation in radiotherapy. Maximum likelihood estimation (MLE) is used to separately determine the measurement error and the random patient position error in a series of external position measurement for 112 prostate patients. Measurement error can be a significant component of the measured variation, with serious implications for IGRT where measurements are used to correct the patient position. Measurements are made of internal and external errors for 44 patients, comparing techniques for rectal and bladder preparation. Internal marker seeds are used to determine prostate position from portal images. Seed based portal imaging is found to be an effective method for IGRT for patients in all arms of the trial. Two techniques for modelling prostate shape are investigated. To analyse patterns of observer variability a method is developed based on the inclusion of image pixels, or voxels, in the shape volume, using principle component analysis (PCA) of binary masks. Inclusion probability is found to be a useful concept but the results of the PCA are poor, demonstrating a high level of dependence between modes. A more conventional approach, the PCA of boundary vectors, is used to generate a 3D deformable vector model of the prostate from correlated 2D orthogonal views. The model was trained from data for 28 patients with 2 CT scans and 2 independent observations of prostate shape per scan. MLE was used to separate the components of shape variation. Of the total variance observed 35% was due to population variation, 17% due to day to day variation and 48% due to observer variation. The final simulation tool allows dose to individual tissue elements to be accumulated over a simulated course of treatment for a treatment population. The tool is used to calculate TCP for a wide range of situations; investigating the impact of observer error, target motion, deformation, margin size, IGRT technique, dose prescription and the use of a high dose boost. The reduction in calculated TCP with margin size is shown to be continuous, with no clear cut-off point. The reduction in TCP with 'inadequate' margins can be modest and easily offset by small increments in dose. This clearly validates the recent trend towards small volume, high dose techniques, even where dose coverage might be compromised. In order to truly determine the optimum technique the simulation would need to be expanded to incorporate the risk of damage to the rectum. Even in its current form, the simulation tool still provides a unique insight into the relationship between the dose-volume prescription, the IGRT technique and the population TCP; allowing an objective assessment of the relative benefits of different approaches to prostate radiotherapy.

Item Type: Thesis (PhD)
Contributors: Harvey, R (Supervisor), Fisher, M (Supervisor) and Ritchings, T (Supervisor)
Schools: Schools > School of Computing, Science and Engineering
Depositing User: Institutional Repository
Date Deposited: 30 Jul 2021 10:37
Last Modified: 27 Aug 2021 21:56

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)