Realizing a new paradigm in radiation therapy treatment planning

This thesis investigates the feasibility of a new IMRT planning paradigm called Interactive Dose Shaping (IDS). The IDS paradigm enables the therapist to directly impose local dose features into the therapy plan. In contrast to the conventional IMRT planning approach, IDS does not employ an objective function to drive an iterative optimization procedure. In the first part of this work, the conventional IMRT plan optimization method is investigated. Concepts for a near-optimal implementation of the planning problem are provided. The second part of this work introduces the IDS concept. It is designed to overcome clinical drawbacks of the conventional method on the one hand and to provide interactive planning strategies which exploit the full potential of modern high-performance computer hardware on the other hand. The realization of the IDS concept consists of three main parts. (1)A two-step Dose Variation and Recovery (DVR) strategy which imposes localized plan features and recovers for unintentional plan modifications elsewhere. (2)A new dose calculation method (3)The design of an IDS planning framework which provides a powerful graphical user interface. It could be shown that the IDS paradigm is able to reproduce conventionally optimized therapy plans and that the IDS concepts can be realized in real-time