Metal artifact reduction algorithm for image quality improvement of cone-beam CT images of medium or large cerebral aneurysms treated with stent-assisted coil embolization.

Abstract

The aim of the present study was to assess image quality improvement using a metal artifact reduction (MAR) algorithm in cases of medium or large cerebral aneurysms treated with stent-assisted coil embolization (SAC), and to analyze factors associated with the usefulness of the MAR algorithm. We retrospectively evaluated the cone-beam computed tomography (CBCT) data sets of 18 patients with cerebral aneurysms treated with SAC. For subjective analysis, images of all cases with and without MAR processing were evaluated by five neurosurgeons based on four criteria using a five-point scale. For objective analysis, the CT values of all cases with and without MAR processing were calculated. In addition, we assessed factors associated with the usefulness of the MAR by analyzing the nine cases in which the median score for criterion 1 improved by more than two points. MAR processing improved the median scores for all four criteria in 17/18 cases (94.4%). Mean CT values of the region of interest at the site influenced by metal artifacts were significantly reduced after MAR processing. The maximum diameter of the coil mass (< 17mm; odds ratio [OR], 4.0; 95% confidence interval [CI], 1.2-13.9; p = 0.02) and vessel length covered by metal artifacts (< 24mm; OR, 2.3; 95% CI, 1.1-4.7; p = 0.03) was significantly associated with the usefulness of the MAR. This study suggests the feasibility of a MAR algorithm to improve the image quality of CBCT images in patients who have undergone SAC for medium or large aneurysms.