The MedApp treatment facility of FRM II provides a horizontally fixed beam of unmoderated fission neutrons. Since high LET radiation can deactivate cells rather independently of their phase in the cell cycle, their repair capacities and their oxygenation state, fast neutrons offer advantages especially for the treatment of radioresistant tumours. Scope of this thesis is the establishment of a treatment planning system (TPS) for the MedApp facility with retrospective purposes. The development of a personalised dosimetric tool will improve the accuracy of the predictions of the dose absorbed by the different tissues surrounding the tumour and by the tumour itself. This will improve the quality of treatments with fast neutrons. The capabilities of different simulation softwares used for that purpose were studied. The main objective of the thesis was to find out whether these simulation programs were suitable to run TPS simulations for fission neutron radiotherapy at the MedApp facility. Two programs, SERA and ImageJ combined with MCNPX, were chosen for a closer analysis. The features of both programs were studied. To match the MedApp requirements their capabilities were extended. So as to facilitate the user task, work was done to adapt these programs and scripts were written in order to generate the inputs data for both programs. The segmenting process which is the most time consuming step in the planning procedure, was analysed in order to find a possible procedure which simplifies the process, especially for novice users. To verify the suitability of the studied programs, calculations of the dose distribution in different phantoms were made and compared with measurements. Moreover, calculations with human voxelised phantoms were also carried out in order to test the suitability of the programs for human dosimetry. The results obtained with both simulations programs were in good agreement with the depth dose measurements in the water phantom of the MedApp facility. Moreover, the simulation times were acceptable. These verifications demonstrated that both computational dosimetry systems have enough performance for fast neutron therapy dosimetry.

Outgoing