PET image super resolution using a novel algorithm in low resolution sinogram reconstruction

Abstract

Medical imaging is a powerful technique to extract a 3D model of the human body or specific areas of affected organs. Computed Tomography(CT), Magnetic Resonance Imaging(MRI) are the three dimensional medical imaging techniques that have been developed earlier . Positron Emission Tomography (PET) is a 3D medical imaging technique wherein radioactive tracers are introduced into the body by either injecting the patient with a radioactive tracer compound or inhalation of radioactive gas by the patient. After injection of the tracer compound labeled with a positron emitting radionuclide the patient for PET study is placed within a PET scanner after which the scan image is acquired. The scan image so generated is of low resolution that makes the visualization of smaller units of neuronal activity difficult hence making medical analysis difficult. A novel and efficient system is needed to improvise resolution of PET scan image since higher resolution is a key factor for increased information content. High resolute image is mandatory for increased accuracy in understanding the morphological structure of organs, for early detection of abnormalities and suspected pathologies in the patient. This paper proposes a better approach to effectively increase the resolution of the PET image by applying Super Resolution (SR) techniques to the low resolution 2D sinograms.