Robust Evaluation and Optimisation for Treatment Planning in High-Dose-Rate Prostate Brachytherapy

Simple item page
dc.contributor.advisorSantos, Alexandre
dc.contributor.advisorDouglass, Michael
dc.contributor.authorKennedy, Andrew Christopher
dc.contributor.schoolSchool of Physics, Chemistry and Earth Sciencesen
dc.date.issued2024
dc.description.abstractUncertainties in high-dose-rate (HDR) prostate brachytherapy (BT) results in a treatment delivered to the patient that is a deviation from what was planned and may fail clinical objectives. Margins around the clinical target volume (CTV) are often used to account for uncertainties; however, these are not patient-specific and may cause high doses to organs at risk or reduced coverage of the CTV. Robust evaluation directly simulates the effect of uncertainties on a nominal plan, quantifying its robustness to uncertainty. Incorporating robustness measures into an optimiser generates treatment plans with maximised robustness to uncertainty. Forty-nine patients were selected at random that had been treated using a live ultrasound planning technique utilising Vitesse (v4.0.1) treatment planning system (TPS). Prostate volumes ranged from 11.8 cc to 68.6 cc and needles ranged from 11 to 21 needles. A robust evaluation algorithm was developed that simulates 14 major uncertainties in the technique by changing six parameter variables from their nominal values. Probabilistic uncertainty scenarios were formed from a random sampling of six probability density functions. Robustness was measured as the pass-rates from 1000 uncertainty scenarios for eight dosevolume- histogram (DVH) metrics, the clinical objectives. The algorithm was explained in detail and comprehensively evaluated on one patient plan. Next, the robustness of 49 patient plans was quantified by calculating three DVH metrics passrates from 5000 uncertainty scenarios. The highest robustness was for the maximum dose to the rectum, D0.1 cc R , with median pass-rate of 96.5% (min: 80.1%, max: 100.0%), while the maximum dose to the urethra, D0.01 cc U , was least robust with pass-rate of 79.0% (min: 44.0%, max: 89.9%). The minimum dose to the hottest 90% of the prostate, D90 P , was 84.9% (min: 34.7%, max: 98.4%). The overall pass-rate, all objectives passing simultaneously, had a median of 63.4% (min: 14.0%, max: 82.5%). Two-dimensional dwell point movements along axial slice directions produced the worst-case scenarios for the three DVH metrics. A robust optimisation method was developed using a genetic algorithm incorporating three objective functions that approximate the pass-rates for D90 P , D0.01 cc U , and D0.1 cc R . Two-hundred robust-optimised plans were generated that were the best trade-off for the three approximate pass-rates. The algorithm was evaluated on one patient, showing a significant increase in robustness with the overall pass-rate increasing from 2.6 ± 1.5% for the TPS-optimised plan to 70.2 ± 1.5% for the best robust-optimised plan. Next, the robustly-optimised plans for 49 patients were compared against their corresponding TPS-optimised plans. The robust-optimised plans had a higher median overall pass-rate of 55.1% (min: 11.6%, max: 71.2%) compared with 35.4% (min: 2.5%, max: 50.6%), a higher D0 U .01 cc passrate of 64.8% (min: 35.0%, max: 94.6%) compared with 49.2% (min: 14.4%, max: 69.2%). The other two DVH metrics showed a slight increase in robustness. The robust evaluation algorithm was shown to quantify the robustness of nominal treatment plans in HDR prostate BT, providing additional information about the likelihood of the treatment passing clinical objectives. The robust optimisation algorithm was able to generate more robust plans on average than the TPS, particularly for plans with low robustness.en
dc.description.dissertationThesis (Ph.D.) -- University of Adelaide, School of Physics, Chemistry and Earth Sciences, 2024en
dc.identifier.urihttps://hdl.handle.net/2440/141588
dc.language.isoenen
dc.provenanceThis electronic version is made publicly available by the University of Adelaide in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material you wish to be removed from this electronic version, please complete the take down form located at: http://www.adelaide.edu.au/legalsen
dc.subjectRobust evaluationen
dc.subjectRobust optimisationen
dc.subjectBrachytherapyen
dc.subjectRadiotherapy planningen
dc.subjectUncertaintyen
dc.titleRobust Evaluation and Optimisation for Treatment Planning in High-Dose-Rate Prostate Brachytherapyen
dc.typeThesisen

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
Kennedy2024_PhD.pdf
Size:
9.83 MB
Format:
Adobe Portable Document Format
Description:
Download

Collections

Research Theses