Does iterative reconstruction improve image quality and reduce dose in computed tomography?

The Filtered Back Projection (FBP) algorithm has been the standard algorithm for image reconstruction in computed tomography (CT) for many years, but it creates possible streak artifacts and a notable increase in image noise when radiation dose is reduced exceedingly. With technological advancement and increased computational capacities in workstations, iterative reconstruction (IR) algorithms have re-emerged as a potential alternative. The purpose of this review is to establish if there is a general consensus that IR algorithms faithfully reduce radiation dose and improve image quality in CT in comparison with the FBP algorithm. A systematic review of the literature from 2012 to 2015 was conducted using Ovid MEDLINE and PubMed databases, as well as various well-known journals such as the American Journal of Roentgenology, European Journal of Radiology, Physica Medica and the Korean Journal of Radiology.
A total of 57 articles were categorized as either synopsis articles or performance evaluation clinical studies, where the latter was further divided into 6 sub-categories according to the type of IR algorithm examined. The results show that the use of IR algorithms reduces objective image noise, and at least preserves spatial resolution and low contrast detectability, even when dose is reduced. The findings are also applicable to specific patient groups, such as pediatrics and obese patients. In conclusion, there is a general consensus that IR algorithms can faithfully reduce radiation dose and improve image quality in CT in comparison with the FBP algorithm.