On lesion detectability by means of 300ps-FWHM TOF whole-body RPC-PET: An experiment-based simulation study

A single-bed, whole-body positron emission tomograph based on resistive plate chamber detectors has been proposed (RPC-PET). It has been shown by simulation that RPC-PET with an axial field-of-view (AFOV) of 2.4m is feasible and yields an absolute sensitivity enhancement of at least one order of magnitude superior to that of typical cylindrical, crystal-based PET scanners. In addition to its time-of-flight (TOF) advantage, RPC-PET offers potential very-high spatial resolution at the detector level. A fully-3D reconstruction algorithm capable of processing the very inclined lines-of-response (LOR) from large AFOV systems such as RPC-PET has been demonstrated. It relies on the application of a TOF-based-kernel into the maximum likelihood estimation maximization algorithm. By means of a 300 ps full width at half maximum (FWHM) time resolution, a rejection of 77% of the scattered events was obtained. It is shown that the scatter fraction rejection grows exponentially with an increasing time resolution. We present reconstructed results from blind simulations corresponding to an anthropomorphic phantom with oncological lesions of several sizes immersed into different locations within the human body. A comparison between 300 and 600 ps FWHM TOF reconstructed images is performed. An increasing detectability is observed for a better TOF resolution. We finally compare issues related to image convergence speed and computational burden, making use of graphical processing units (GPUs) and 16 threads central processing units (CPUs). GPUs perform better by a factor two in speed. An alternative approach, which consists in dividing the acquired data into five different image regions, which are reconstructed independently, provides a three times faster reconstruction, as compared with whole-body reconstruction, and allowing to reach a reconstructed image by means of a 300ps FWHM RPC-PET scanner in 7 minutes after the end of data acquisition.