Truncation artifact correction for micro-CT scanners.

The work included in this project is framed on one of the lines of research carried out at the Laboratorio de Imagen Médica de la Unidad de Medicina y Cirugía Experimental (UMCE) of Hospital General Universitario Gregorio Marañón and the Bioengineering and Aerospace Department of Universidad Carlos III de Madrid. Its goal is to design, develop and evaluate new data acquisition systems, processing and reconstruction of multimodal images for application in preclinical research. Inside this research line, an x-ray computed tomography (micro-CT add on) system of high resolution has been designed for small animal. Nowadays, computed tomography (CT) is one of the techniques most widely used to obtain anatomical information from living subjects. Different artifacts from different nature usually degrade the qualitative and quantitative analysis of these images. This creates the urgent need of developing algorithms to compensate and/or reduce these artifacts.

The general objective of the present thesis is to implement a method for compensating truncation artifact in the micro-CT add-on scanner for small animal developed at Hospital Universitario Gregorio Marañón. This artifact appears due to the acquisition of incomplete x-ray projections when part of the sample, especially obese rats, lies outside the field of view. As a result of these data inconsistencies, bright shading artifacts and quantification errors in the images may appear after the reconstruction process.

First of all, truncation artifact in the high resolution micro-CT add-on scanner was studied. Then, after a review of the proposed methods in the literature, the optimal approach for the micro-CT addon was selected, based on a sinogram extrapolation technique developed by Ohnesorge et al [1]. This method consists on a symmetric mirroring extrapolation of the truncated projections that guarantees continuity at the truncation point. It includes a sine shaping effect that ensures a smooth attenuation signal drop. Truncation artifact correction method has been validated in simulated and real studies. Results show an overall significant reduction of truncation artifact. This algorithm has been adapted and implemented in the reconstruction interface of the preclinical high-resolution micro-CT scanner, which is manufactured by SEDECAL S.L. and commercialized worldwide.