Verification of patient position for proton therapy using portal X-Rays and digitally reconstructed radiographs

[摘要] This thesis investigates the various components required for the developmentof a patient position verification system to replace the existing system usedby the proton facilities of iThemba LABS1. The existing system is basedon the visual comparison of a portal radiograph (PR) of the patient in thecurrent treatment position and a digitally reconstructed radiograph (DRR)of the patient in the correct treatment position. This system is not only oflimited accuracy, but labour intensive and time-consuming. Inaccuracies inpatient position are detrimental to the effectiveness of proton therapy, andelongated treatment times add to patient trauma. A new system is neededthat is accurate, fast, robust and automatic.Automatic verification is achieved by using image registration techniquesto compare the PR and DRRs. The registration process finds a rigid bodytransformation which estimates the difference between the current positionand the correct position by minimizing the measure which compares thetwo images. The image registration process therefore consists of four maincomponents: the DRR, the PR, the measure for comparing the two imagesand the minimization method.The ray-tracing algorithm by Jacobs was implemented to generate the DRRs,with the option to use X-ray attenuation calibration curves and beam hardeningcorrection curves to generate DRRs that approximate the PRs acquiredwith iThemba LABS's digital portal radiographic system (DPRS)better.Investigations were performed mostly on simulated PRs generated from DRRs, but also on real PRs acquired with iThemba LABS's DPRS.The use of the Correlation Coefficient (CC) and Mutual Information (MI)similarity measures to compare the two images was investigated.Similarity curves were constructed using simulated PRs to investigate howthe various components of the registration process influence the performance.These included the use of the appropriate XACC and BHCC, thesizes of the DRRs and the PRs, the slice thickness of the CT data, theamount of noise contained by the PR and the focal spot size of the DPRS'sX-ray tube.It was found that the Mutual Information similarity measure used to compare10242 pixel PRs with 2562 pixel DRRs interpolated to 10242 pixelsperformed the best. It was also found that the CT data with the smallestslice thickness available should be used. If only CT data with thick slices isavailable, the CT data should be interpolated to have thinner slices.Five minimization algorithms were implemented and investigated. It wasfound that the unit vector direction set minimization method can be usedto register the simulated PRs robustly and very accurately in a respectableamount of time.Investigations with limited real PRs showed that the behaviour of the registrationprocess is not significantly different than for simulated PRs.

[发布机构] Stellenbosch University