Abstract
OBJECTIVE: To evaluate the dose efficiency of cadmium-zinc-telluride (CZT) based photon-counting CT (PCCT) compared to energy-integrating detector CT (EID-CT) across phantom sizes. METHODS: A patient-specific 3D-printed pancreas phantom and a phantom with tissue mimicking inserts were placed in extension rings corresponding to the 50(th), 75(th), 85(th), and 95(th) percentile adult waist circumferences. Phantoms were scanned on both PCCT and EID-CT with CTDI(vol) ranging from 0.5 to 19.4 mGy. Noise was measured in both phantoms to evaluate dose efficiency. Non-Poisson noise at low doses (<2 mGy) was quantified using root mean square error from linear fits of the noise-dose relationship. Potential dose reduction was then assessed by matching noise levels between scanners across phantom sizes. RESULTS: PCCT demonstrated reduced noise compared to EID-CT across all phantom sizes and doses with average noise reductions of 22%, 23%, 25%, and 28% for the 50(th), 75(th), 85(th), and 95(th) percentile phantoms, respectively. Noise reduction intensified at lower doses and larger phantom sizes, reaching 88 HU at 1 mGy for the 95(th) percentile phantom. Non-Poisson noise decreased significantly with PCCT compared to EID-CT for all phantom sizes (p < 0.013). At matched noise levels, PCCT enabled dose reductions of 33% and 44% for the 50(th) and 95(th) percentile phantoms, respectively. CONCLUSIONS: PCCT exhibited superior dose efficiency compared to EID-CT across a range of phantom sizes. The enhanced dose efficiency enables both noise reduction and potential dose reduction for the imaging of obese patients and low-dose imaging applications.