Sparse Detector Configuration in SiPM Digital Photon Counting PET: a Feasibility Study

To investigate the minimum number of SiPM detectors required for solid-state digital photon counting (DPC) oncologic whole-body 2-deoxy-2-[¹⁸F]fluoro-D-glucose ([¹⁸F]FDG) positron emission tomography (PET)/X-ray computed tomography (CT).

A DPC PET/CT (Vereos, Philips) with 23,040 1-to-1 crystal-to-detector couplings was utilized. [¹⁸F]FDG PET/CT of a uniformity phantom and 10 oncology patients selected by block randomization from a large clinical trial were included (457 ± 38 MBq, 64 ± 22 min p.i, body mass index (BMI) of 14-41). Sparse-ring PET configurations with 50 % detector reduction in tangential and axial directions were analyzed and compared to the current full ring configuration. Resulting images were reviewed blindly and quantitatively over detectable lesions and the liver.

One hundred twelve lesions (d = 10 to 95 mm) were analyzed in the patient population. All lesions remained visible and were demonstrated without compromised image quality under all BMIs in the 50 % sparse detector configurations despite the DPC PET system sensitivity reduction to 1/4th. An excellent consistency of SUV_max measurements of lesions with an average of 5 % SUV_max difference was found between dPET of full and sparse configurations.

The feasibility of either expanding the axial field of view (FOV) by a factor of two or halving the number of detectors was demonstrated for solid-state digital photon counting PET, thus either potentially enabling cost reduction or extended effective axial FOV without increased cost.