3D Artifact Localization Using Connected Components

Medical imaging is a useful tool that simplifies the diagnosing and treatment. Having high-quality images doctor can make a statement about disease and correct the further recovery steps. The image quality depends on such factors as equipment condition, scanning correctness and patient’s body structure. The solid inclusions inside the human body cause the “dark and bright streaks at computed tomography (CT)” (Katsura et al. in Radiographics 38:450–461, 2018). This work aims to reduce the effect of the metal body artifacts in the computed tomography (CT) automatically. Artifacts are parts of the image that corrupt information—it changes the pixel intensities. Proposed technique localizes artifacts as the first step of the artifact removal algorithm. Artifacts appear in CT images as 3D connected components. The artifact’s area represents a connected component the within a stack of binarized images. The Gaussian mixture model of pixel intensity histogram allows to estimate the adaptive binarization threshold. Describing artifact seпmentation method increases the Jaccard score from 0.247 to 0.996 and from 0.561 to 0.943 for 2 data cases compared to static global thresholding.