Effective Radiation Dose Reduction in Computed Tomography With Iterative Reconstruction in Patients With Urinary Stone.

Abstract

This study aimed to prospectively compare the image quality and visibility of urinary stone on computed tomographic (CT) images at multiple radiation exposure levels from the same patient reconstructed with sinogram-affirmed iterative reconstruction (SAFIRE). This study included 760 patients with urinary stone who underwent CT with simultaneous acquisition at 6 exposures per patient (100% filtered back projection, 75%, 50%, 37.5%, 25%, and 12.5% SAFIRE). Two radiologists independently assessed overall image quality, noise, and stone visibility by using a 5-point scale. Quantitative measurements, including the CT number, image noise, signal-to-noise ratio, contrast-to-noise ratio (CNR), and corresponding figure of merit (FOM), were compared for 100% versus 5 other radiation doses. Qualitative overall image quality, noise, and stone visibility according to the location were not inferior at 37.5% exposure compared with 100% exposure, except for the visualization of smaller stones <3 mm. The signal-to-noise ratio and CNR of CT images were increased at 50% exposure compared with 100% exposure. Computed tomographic images at 37.5% exposure reconstructed with SAFIRE had significantly more noise and a lower CNR compared with CT images reconstructed with filtered back projection, based on FOMnoise and FOMCNR. The size-specific dose estimation was 4.1 ± 0.8 mGy at 37.5% exposure. Computed tomography performed at 37.5% exposure with SAFIRE may be diagnostically acceptable for the detection of clinically relevant stone.