Publikationsrepositorium - Helmholtz-Zentrum Dresden-Rossendorf

In-beam positron emission tomography (PET) with a double-head PET scanner has been successfully applied for the in situ dose monitoring of about 440 patients with static tumour entities within a pilot project at the GSI Helmholtzzentrum für Schwerionenforschung in Darmstadt, Germany. For future treatment of intra-fractionally moving tumours, mainly located in the lung or liver, a dose monitoring is highly desired since complex beam delivery strategies and continuous respiration-related density changes in the irradiated tissue increase the risk for dose errors. Conventional three-dimensional (3D) in-beam PET measurements taken from moving structures would result in a reproduction of a blurred activity distribution and would, therefore, impede correct dose verification. An implemented motion compensated (4D) reconstruction routine based on the existing 3D Maximum Likelihood Expectation Maximization (MLEM) algorithm has been tested by phantom experiments performed at the GSI facility. Homogeneous polymethyl methacrylate (PMMA) phantoms of dedicated geometries have been irradiated with 12C pencil beams and in-beam PET data acquisition was performed throughout irradiation and additional time beyond to gain better statistics. Motion mitigation techniques of rescanning, gating or tracking have been used for a homogeneous dose deposition in the PMMA targets which performed a one-dimensional cos2- or cos4-shaped motion perpendicular to the beam direction. 4D reconstruction results have been compared to static reference measurements and to reconstruction results without consideration of the
target motion. Evaluation outcome with respect to the conservation of lateral and distal gradients in the reconstructed activity distributions will be shown at the workshop. Furthermore, the acquisition of required input parameters from the accelerator, the motion acquisition and the motion compensation systems and the data handling before reconstruction, like the temporal synchronization of all signals, as well as the necessary improvements for clinical usage of the 4D in-beam PET dose monitoring will be discussed.