Purpose: To compare image quality at low- and routine-dose between a photon-counting (PCCT) and an energy integrating (EICT) CT system at different patient sizes.
Methods: A size-variant phantom (Mercury 3.0) was used to characterize image quality as a function of patient size. The phantom contained five cylinders of different diameters, each with two sections: one uniform and the other with air, water, bone, polystyrene, and iodine inserts. The phantom was scanned on a Siemens SOMATOM Definition Flash and NAEOTOM Alpha at two dose levels of 1.18 mGy (low) and 7.51 mGy (routine) without tube current modulation and without changing other acquisition settings. FBP reconstructions were performed with different Br kernels. NPS, TTF, CNR, and detectability index (d’) for a task of identifying 5 mm disk were evaluated based on AAPM TG-233 metrology using imQuest, an open-source software package.
Results: For a given dose level, phantom diameter, and resolution, the noise was lower while the contrast, CNR, and d’ were higher or comparable for PCCT. The CNR difference between two systems increased with diameter from about 4% (120 mm) to 50% (370 mm) for low- and about 2% (120 mm) to 30% (370 mm) for routine-dose. For low-dose, d’ difference increased from -1% (120 mm) to 7% (370 mm) while were comparable for routine dose. The percent differences varied with the choices of kernels.
Conclusion: For similar sharpness level, dose to the patient can be significantly reduced by using PCCT to achieve similar noise and contrast in the image as compared to EICT. At both dose levels, PCCT performed better than EICT, but the performance difference was more significant at low- than at high-dose. For a given dose level, the performance difference increased with diameter. In general, PCCT performed better than EICT at both dose levels and for all phantom sizes.
Image Analysis, MTF, Noise Power Spectrum