Title :
Systematic investigation of the reconstruction of images from transmission tomography using a filtered backprojection and an iterative OSML reconstruction algorithm
Author :
Ziegler, A. ; Heuscher, D. ; Köhler, Th ; Nielsen, T. ; Proksa, R. ; Utrup, S.
Author_Institution :
Philips Res. Lab., Hamburg, Germany
Abstract :
The reconstruction of 2D images from fan-beam transmission measurements is performed with filtered-backprojection (FBP) and an ordered subsets maximum likelihood (OSML) method. A special software phantom, which allows measuring the resolution and noise in a non-ambiguous way, is used to simulate transmission tomography scans with different signal-to-noise ratios. The noise and modulation transfer function is calculated for both FBP and OSML reconstruction at several positions distributed over the field of view (FOV). The OSML method is able to achieve a nearly homogeneous high resolution over the whole field-of-view, which is not possible with FBP, because the OSML reconstruction from the focus-centered projections intrinsically uses the improvement of the resolution due to the quarter detector offset without any extra re-binning. In addition, the reconstruction with OSML has a higher resolution than FBP at a similar level of noise, or equivalent, the OSML method achieves a lower level of noise compared to FBP at the same resolution. The reconstruction with OSML can save a factor of about 2 of dose with respect to FBP.
Keywords :
computerised tomography; image reconstruction; image resolution; iterative methods; maximum likelihood estimation; medical image processing; phantoms; 2D images; fan-beam CT scan; fan-beam transmission measurements; field of view; filtered backprojection; focus-centered projections; image reconstruction; image resolution; iterative OSML reconstruction algorithm; modulation transfer function; ordered subsets maximum likelihood method; quarter detector offset; signal-to-noise ratios; software phantom; transmission tomography scans; Image reconstruction; Imaging phantoms; Maximum likelihood detection; Noise level; Noise measurement; Performance evaluation; Reconstruction algorithms; Signal resolution; Signal to noise ratio; Tomography;
Conference_Titel :
Nuclear Science Symposium Conference Record, 2004 IEEE
Print_ISBN :
0-7803-8700-7
DOI :
10.1109/NSSMIC.2004.1462766