Development of Scalable Streaming Readout Data Acquisition and Analysis Platform

Total-body positron emission tomography (TB-PET) requires coincidence-pair event analysis within near-real time from thousands of individual detector modules. Shifting signal digitization closer to the detector hardware enables the use of software-based event analyses and coincidence triggers to handle increasingly complex configurations, in what is called streaming readout (SRO) data acquisition (DAQ) and analysis. Recent developments in nuclear physics detector system front-end design and analysis systems advanced at the Thomas Jefferson National Accelerator Facility (JLab) can support the expected detector modularity and data rate needs of TB-PET. To demonstrate the scalability of TB-PET capable detectors and analysis systems utilizing open-source software and commercially available detector components, in this study we propose a scalable SRO DAQ and analysis architecture, demonstrated with a prototype pre-clinical PET platform. The platform comprises two detector arms with four modules each. Each module consists of an LYSO crystal scintillator array read out by a pixelated SiPM, whose signal is amplified by a commercial PETIROC ASIC and passed to a single FPGA one ach arm for event processing. The event analysis, including the application of all calibrations is done by analysis processes managed by the Environment for Real-time Streaming, Acquisition and Processing (ERSAP) system developed at JLab. We investigate and show the design of modular detector front-end systems, the ERSAP-enabled distributed SRO analysis system, and initial PET imaging results from pre-clinical scans performed with proof-of-principle prototypes at the University of Maryland School of Medicine.