A 64-pixel positron-sensitive surgical probe

We report on the continued development of a 64-pixel positron-sensitive surgical probe with a dual-layer detector and a multi-anode PMT. An 8 /spl times/ 8 array of thin plastic scintillators in the first layer detects positrons and a matched GSO crystal array in the second layer detects annihilation 511 keV gammas, which are required to be in coincidence with the detected positrons. Also, the 64 PMT anode signals are differentiated and an overshoot threshold is applied to separate the fast decay plastic anode signals from the slower GSO anode signals. Finally, an energy threshold is applied to the summed anode signal to distinguish 511 keV gammas from the 140 keV gammas commonly used in sentinel lymph node (SLN) surgery. Previously we reported on how these signal selection criteria were individually tested and optimized based on 9 channels of prototype electronics. Currently the electronics have been upgraded to Xilinx/spl reg/ programmable components, allowing on-the-fly alteration of signal selection criteria, and all 64 channels are operational. Initial measurements of the complete 64-pixel probe were conducted using /sup 18/F-FDG positron sources and /sup 18/F-FDG and /sup 99m/Tc phantoms (background 511 keV and 140 keV gammas), simulating lesions in the SLN surgery environment. The average positron sensitivity is measured to be 3.0-7.0 kcps//spl mu/Ci at different signal selection criteria. The lower bound on sensitivity corresponds to settings optimized for high image resolution and high background rejection ability. The upper bound on sensitivity corresponds to settings optimized for high sensitivity at the cost of lower image resolution and lower background rejection ability. The measured true-to-background contrast in the presence of clinically observed levels of 511 keV and 140 keV background gammas is /spl sim/3:1 for a tumor-to-background uptake ratio of 5:1. Performance measurements of the complete 64-pixel probe including sensitivity, true-to-background ratio, and the pixel separation ability are presented.