Depth-of-Interaction and Coincidence Time Resolution Characterization of Ultrahigh Resolution Prism-PET Modules

Introduction: We characterize the depth-of-interaction (DOI) and coincidence-time-resolution (CTR) performances of Prism-PET modules, the first PET modules to use our unique prism light guide array that achieves comparable performance to traditional dual-ended readout DOI-PET modules. In this work, we present our DOI tagging experiment and CTR experiment, as well as the performance analysis on energy weighted average scheme and time-of-flight (TOF) readout.

Methods: We conducted DOI characterization tests using coincidence events detection, where the line-of-response (LORs) were formed using a lead cylinder with 1 mm diameter pinhole on both ends, and a Na-22 source (3 MBq, 1 mm active diameter) placed in the center. A 4-to-1 coupled Prism-PET module was used as the reference along with three separate test modules: one 4-to-1 coupled uniform light guide PET module, one 4-to-1 coupled Prism-PET module and one 9-to-1 coupled Prism-PET modules. All three 4-to-1 modules consisted of an array of 16 x 16 segmented 1.4 x 1.4 x 20 mm³ LYSO crystals coupled 4-to-1 on one end to 8 x 8 MPPC readout arrays with 3.2 x 3.2 mm² silicon photomultiplier (SiPM) pixels. The 9-to-1 coupled Prism-PET module had a 24 x 24 multicrystal scintillator array with 0.96 x 0.96 x 20 mm³ LYSO crystals coupled 9-to-1 on one end to an 8 x 8 MPPC readout array with 3.2 x 3.2 mm² SiPM pixels. Alignment between reference and test modules was implemented via laser beam through the lead pinhole. The reference crystal (on the reference module) used for coincidence DOI collimation was identified by visual inspection. All crystals in each module were fully polished and the intercrystal spaces were filled with BaSO₄. To acquire the DOI tagging data with precise LOR positioning, the test module was placed on the other side of the lead and raster scanned by using an automated XZ stage with an electrical motor. DOI for each event was estimated by energy weighted average method and the DOI resolution at each depth was determined based on the width of the Gaussian distributions. The CTR measurement was performed using virtually the same configuration as the DOI characterization, but with the reference module rotated 90 degrees so that it was in the same orientation as the DOI test module. All experiments were performed in a temperature-controlled chamber at ~ 20℃ . Data acquisition and readout were done with the PETsys TOFPET2 ASIC Evaluation Kit.

Results: The 9-to-1 Prism-PET module has improved DOI resolution (3.5 mm FWHM) over the 4-to-1 coupled uniform glass module (5 mm FWHM) but has slightly degraded DOI resolution compared to the 4-to-1 Prism-PET module (2.5 mm FWHM). The TOF data analysis shows CTR ~ 260ps for the 4-to-1 Prism-PET module and CTR ~ 280ps for the 9-to-1 Prism-PET module. Conclusion: This research demonstrated two important characterizations and several other features of our ultrahigh resolution Prism-PET modules. The results of our DOI and CTR experiments show great potential for our 9-to-1 Prism-PET modules for future PET imaging systems. Acknowledgments: We would like to thank our colleagues at PETsys Electronics SA. Figure 1. (a) DOI experiment configuration. (b) 9-to-1 DOI test module collimated flood histogram. (c) 4-to-1 and 9-to-1 Prism-PET module DOI Comparison. (d) 4-to-1 module best CTR scatter plot. (e) 9-to-1 module best CTR scatter plot.

• Download figure
• Open in new tab
• Download powerpoint