Artifact meaning xray12/22/2023 However, most of this work is restricted to conventional energy-integrating CT applications. In the field of X-ray CT imaging, several studies using deep learning methods have shown promising results in MAR, scattering correction, ring artifact reduction, and reconstruction in low-dose regimes. Convolutional neural network (CNN) architectures have shown a lot of success recently in fields such as computer vision and image classification and segmentation. However, these traditional correction methods typically require a high amount of prior knowledge for iterative regularization, are tailored ad-hoc for certain applications, and fail to offer a generalizable solution. Alternatively, MAR can be conducted in a post-reconstruction step, by directly attempting to remove the streaks and restore the LAC values of the materials. This is typically conducted by adaptive interpolation of the photon-starved detector pixels in the projections. This can be conducted in a pre-reconstruction approach by applying a correction algorithm in the sinogram domain. The goal of a MAR algorithm is to improve the quality of the reconstructions by removing the streaks while preserving the real features. For these reasons, metal artifact reduction (MAR) methods are key to the accurate and precise performance of a CT scanner both in medical and non-destructive testing fields. The metal artifacts not only degrade the graphic quality of the reconstructions, but also additionally challenge the quality of segmentation and thus, the classification and characterization of materials. This effect yields severe streaking artifacts in the reconstructions that typically originate from a metallic object and extend further, overlapping with the other objects or materials present in the volume. However, this technique still suffers from degrading effects, such as photon starvation, caused by the presence of high attenuation materials (e.g., metals). It has been demonstrated that using this technique yields enhancement of the contrast-to-noise ratio (CNR), the characterization of materials and the discrimination of threat objects. In SCT, the energy dependence of the linear attenuation coefficient (LAC) of materials is measured with the aid of detectors able to discriminate the energy of the incoming photons in a discrete number of energy channels. Spectral X-ray computed tomography (SCT) is an emerging technique for enhanced non-destructive investigation of inner features of objects. The correction method is parameter free and only takes around 15 ms per energy channel, satisfying near-real time requirement of industrial scanners. We show that the additional information in the energy domain provides relevance for restoring the quality of low-energy reconstruction affected by metal artifacts. The correction efficiently reduces streaking artifacts in all the energy channels measured. We present a correction method for metal artifact reduction in SCT that is based on spectral deep learning. While this mitigates energy-dependent distortions such as beam hardening, metal artifacts due to photon starvation effects are still present, especially for low-energy channels where the attenuation coefficients are higher. Compared with the conventional X-ray CT, this technique provides spectral photon energy resolution in a finite number of energy channels, adding a new dimension to the reconstructed volumes and images. Spectral X-ray computed tomography (SCT) is an emerging method for non-destructive imaging of the inner structure of materials.
0 Comments
Leave a Reply.AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |