CZT pixel detectors for improved SPECT imaging

CZT pixel detectors show interesting capabilities for SPECT imaging. They combine a high energy resolution, good detection efficiency and a very good intrinsic spatial resolution. As these performances are superior to those obtained with scintillation detectors, a resulting improvement of image quality is expected. However, SPECT imaging performances are mostly limited by the collimation geometry and CZT based imagers cannot bring a significant gain without a global system redesign. Recently, several innovating ideas have been introduced and new systems have been developed in order to overcome these limitations. We previously proposed an architecture named “HiSens” with the objective to use more completely the information given by the detector to improve image quality. We presented some encouraging simulation results showing that a good knowledge of the interaction location inside of the detector could help to improve image. We present here an experimental study using a dedicated test bench called “Minigami”. It includes a monolithic pixel detector associated with integrated charge preamplifiers and 16 fully digital processing channels. Wide images are obtained by scanning the object with a translation table. Signal processing circuits extract relevant parameters relative to each photon. These data are saved to a list mode file. Afterwards acquisition files are post-processed in order to build spectra and images. In this paper, we describe steps involved in the image construction and solutions used to compensate for various detector non-idealities. Finally, we will discuss image characterization and present some results. Results show that CZT intrinsic spatial resolution enables to retrieve additional information and allows to shift the traditional trade-off between efficiency and spatial resolution.