The Formation of Computed Tomography Images from Compressed Sampled One-dimensional Reconstructions

Abstract

Compressive Sensing (CS) algorithms are widely adopted for the reconstruction of Magnetic Resonance images (MRI). Owing to differences in the nature of the measurements acquisition processes, these techniques are still not often employed for X-ray Computed Tomography (CT) imaging. However, CS has the potential of reducing the amount of emitted radiation during the CT acquisition process. This study establishes a structure, based on one-dimensional reconstructions, to build CT images using numerical optimization with direct methods, as opposed to traditional indirect methods, such as Conjugate Gradient. The structure was evaluated with regard to its ideal measurements and obtained better results, in terms of signal-to-noise ratio, with respect the reconstruction based on a Filtered Back Projection (FBP) algorithm.