Analyzing the stability of 256 APDs through leakage current and temperature monitoring in a 1 mm<sup>3</sup> resolution clinical PET system

Abstract

A 1 1 mm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> resolution, high-sensitivity PET camera dedicated to breast imaging is being constructed with 4608 8×8 arrays of 1mm3 LYSO crystals, each coupled to a Position Sensitive Avalanche Photodiode (PSAPD). A subsection of the proposed camera with 256 PSAPDs in each head of the system has been constructed. The gain, as well as leakage current, of each of the PSAPDs shows a strong dependence on both bias voltage and temperature. To monitor these parameters, circuitry to measure leakage current from each of these PSAPDs using a single +5 V supply has been developed along with a USB-based communication protocol to display these values in quasi-real time within a Qt interface. A thermal regulation technique is also presented for the densely packed PSAPDs, along with a test for the long-term performance of this regulation scheme when the APDs are reversed biased at -1700V. Through the use of water-cooled heat sinks, and thermoelectric elements for thermal regulation, the temperature within the PSAPDs, as measured by a thermistor, is held to a range of 0.67 °C (between 18.20 °C and 18.87 °C) over a time period of 7.5 h under steady state operating conditions. This temperature change is smaller than the ambient temperature range of 0.94 °C (between 22.50 °C and 23.44 °C). During the same period, the total leakage current from 256 PSAPDs exhibits only a slight change of -0.206 13±0.000 25 μA h <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-1</sup> (-0.296 06 ± 0.000 01 %/h) showing the system is capable of operating at a consistent level for long durations. The propagation of thermal changes within the camera head is also studied. Modules are arranged in rows of 16. Due to the edge placement of the cooling elements, an average delay of 183±11 s is measured for modules in the center of these rows compared to those at the edge for a change in the thermal regulation to take effect.