Review of image reconstruction techniques in medical transaxial computed tomography

After a brief recall about the basic principle of transaxial computed tomography this paper deals with the main image reconstruction techniques usable in different medical fields when projections are pre-processed according to the kind of detected signals. The analytical techniques which are based on the back-projection operation and deconvolution filtering are distinguished from the algebraic techniques which involve an iterative process for computation of the correction coefficients to be applied to the pixel values. For each technique the reconstruction algorithm is described and an illustration of the data processing steps in the case of a single radioactive source is given. The application conditions and related problems are discussed in the field of single photon emission computed tomography. The main deconvolution filters used in analytical techniques are briefly described and this review is completed by a comparison between the characteristics of analytical and algebraic techniques