Fast Method to Reconstruct and Enhance CT Images: Applied to Metal Artifact Reduction

In the presence of metallic implants in the human body, metallic artefacts are introduced into X-ray CT images. Metallic artefacts frequently lead to heavy artefacts which can obscure important information and could seriously degrade the quality of the CT image, therefore, influence the clinician's diagnostic results. Many methods have been developed in recent years where the problem of reducing the metallic artifact has been widely studied, providing a number of solutions. In this work, we propose a new, fast and efficient method to remove artifacts. The proposed method consists in using the image reconstructed from a sinogram affected by the artefacts in order to generate a synthesized sinogram instead of the originally measured sinogram. The main steps are: the segmentation method, the method of filling gaps in the sinogram and improving the image respectively. We used a fast segmentation method based on the K-means classification. For the recovery of the lost data, we proposed the adaptation of a method based on the discrete cosine transform (DCT). Finally for the reconstructed image enhancement we used an improved contrast equalization technique in order to restore the image intensities to their natural dynamic range. The proposed method has been verified for simulations on a Shepp-Logan phantom as well as on clinical data. This method offers a remarkable improvement in image quality