Study of compressed sensing application to low dose computed tomography data collection

Computed Tomography (CT) is one of the most used imaging techniques in medical field which enables us to generate two-dimensional as well as three-dimensional image of the inside the body. Fundamentally, any imaging system contains two main stages the data collection and the image reconstruction. For CT, The data collection stage depends on the resolution of image collected and field of view. The image reconstruction of an object depends on projections by passing a series of rays through an object. The problem in general is that CT is computationally very intensive, due to the large number of projections. The large computational requirements have led to large times for CT image reconstruction, and extra X-ray dose to get high quality images. To accelerate the reconstruction process and decrease the effect of x-ray on patients, we need to decrease the number of projections. In this paper we introduce a compressed sensing (CS) technique for CT to reconstruct images from reduced projection data and compare it with other algorithms, A CS iterative algorithm reconstructed an image of digital Shepp-Logan phantom using small number of projections with high quality resolution compared to traditional iterative technique, we studied the effect of algorithm controlling parameters on the reconstructed image. Finally using the proposed technique will decrease the high risk associated with the high dose x-ray needed in the traditional CT scans.