Tomosentezin diş“ x-ray görüntülerine uygulanmasi

Digital tomosynthesis is a method of reconstructing any number oftomographic planes, by using a set of limited angle projections,acquired as the X-ray source moves in an arc around the object.In some cases the quality of the reconstructed planes is affectedby structured noise, due to blur from planes other than thefulcrum plane. The purpose of this project is to implement theMultiple Projection Algorithm along with the out-of-plane blurremoval method that could remove the noise from the reconstructedplanes and improve their quality.For this purpose, one software phantom was used to evaluate theparameters that affect the image quality. Different groups werecreated with varying arc lenghts and projection counts . Theseimages were reconstructed with the Multiple ProjectionsReconstruction Algorithm (MPA) written in MATLAB.Blur removal method was implemented to increase image quality.Image quality was assessed by evaluation of the reconstructedplanes in terms of both visual assessment and SSIM analysis. TheMPA followed by the proposed noise removal algorithm resulted inless noisy tomosynthesis images.The algorithm was also tested with dental phantom data set. Theresults showed that the tomosynthesis image reconstruction is apromising and a low cost method to evaluate digital dental x-rayswithout excess radiation to the patient.