Comparison of image quality in pediatric head computed tomography reconstructed using blended iterative reconstruction versus filtered back projection

We compared image quality in 44 pediatric head computed tomographic exams reconstructed using four levels of blended iterative techniques (iDOSE 1-4) versus filtered back projection. Three radiologists reviewed the 220 series in a randomized, blinded manner. They scored each series based on gray-white matter differentiation, visibility of the pre-pontine cistern, caudate head, image noise, and image quality. Noise was measured in the vitreous and the thalamus. Composite scores were highest with iDOSE 4 (P< .0001). Interobserver agreement was fair to moderate. Spearman's rho of the vitreous (− 0.62, P< .001) and thalamus (− 0.58, P< .001) confirmed a significant inverse correlation between iDOSE level and noise.