Conventional radiation treatment concepts are using one treatment plan for the whole course of irradiation. This plan is achieved by creating a "snapshot" of the patient using CT images. To overcome possible anatomic variations safety margins are added around the gross tumor volume.
Advances in radiation therapy led to image-guided radiation therapy (IGRT), which utilizes actual images of the patient right before treatment. Therefore anatomic movements can be compensated effectively and safety margins can be minimized. However, anatomic deformations can not be compensated. Further progress in imaging techniques enabled the representation of day-to-day variations in the anatomy of patients and introduced the concept of adaptive image guided radiotherapy. There plans are recalculated according to daily anatomic changes inside the patient, including deformations. Recently the idea of using cone-beam CT (CBCT) images not only for for verication of the patient position and monitoring of the anatomic changes, but also for the treatment planning became popular. The ultimate goal would be the use of CBCT images for adaptive planning and dosecalculation to provide the possibility of real time planning right before treatment takes place. This would provide a tool to adapt all types of anatomical changes inside the body, including movement and deformation. Therefore safety margins around tumor tissue could be dramatically reduced or even removed without harming healthy tissue at all. The aim of this thesis is to investigate the applicability of different approaches for dose calculation based on CBCT images. Therefore four different adjustment techniques are used: two conversion curve-based and two non-conversion curve-based techniques.
Dose calculation results are compared to results of planning CT and patient group specic conversion curve- based technique, described previously in literature. In detail, dose distributions are calculated by a commercial Monte Carlo algorithm. The outcome of dose calculations is analyzed by comparison of dose-volume histograms (DVH) and gamma evaluation. The outcome will be discussed concerning the accuracy of the methods compared to CT-based planning and applicability in clinical practice. Differences below 3% for PTVs and OARs compared to planning CT demonstrate reasonable accuracy of the dose calculation for CBCT-based methods together with the required sensitivity to detect anatomic changes in the IGART process. Especially a simple approach using only 3 densities, showed high accuracy combined with a high clinical applicability, due to the small workload required.