Random coincidences during in-beam PET measurements at microbunched therapeutic ion beams

At the experimental carbon ion tumour therapy facility at GSI Darmstadt, in-beam positron emissions tomography (PET) is used to monitor the dose delivery precision. A dual head positron camera has been assembled from commercial detector components in order to measure the β+-activity, induced by the irradiation, simultaneously to the dose application. Despite the positive clinical impact, the image quality is limited by the low counting statistics, orders of magnitude below that in standart PET applications to nuclear medicine. This paper investigates the origin for the noisy acquisition during particle extraction from the synchrottron of GSI. The results demonstrate the failure of standard random correction techniques due to a γ-ray background correlated in time with the carbon ion beam microstructure. This prevents the use of data acquired during beam extraction for imaging. The loss of counting statistics is expected to rise further at the future hospital-based facility at Heidelberg, due to a more efficient utilisation of the accelerator resulting in shorter beam pauses and a reduced treatment time. In respect, this paper provides the basis for a new data acquísition concept tailored to the unconventional application of in-beam PET imaging to therapy monitoring at radiofrequency pulsed radiation sources.