Robust treatment planning for proton therapy: Computational methods to deal with geometric and target volume uncertainties

The dose concentration at the end of the proton range has the potential to significantly reduce exposure of healthy tissues to radiation, while maintaining the curative dose to the tumor. However, the beneficial effect of protons comes at the cost of increased sensitivity to treatment uncertainties, which can lead precisely to over-exposure of healthy tissues. The main objective of this thesis is the development of treatment planning methods that ensure robustness, while avoiding the possible loss of plan quality at the moment of delivery. The focus lies on establishing sound confidence levels for treatment planning and the integration of target volume uncertainty, previously unconsidered in the planning process. Overall, the concepts developed in this thesis regarding robust dose optimization and statistical analysis, offer better tools for managing the trade-offs and robustness of treatments. This can bring us one step closer to delivering on promises made in proton therapy.(BIFA - Sciences biomédicales et pharmaceutiques) -- UCL, 202