Fourier-based approach to interpolation in single-slice helical computed tomography

It has recently been shown that longitudinal aliasing can be a significant and detrimental presence in reconstructed single-slice helical computed tomography (CT) volumes. This aliasing arises because the directly measured data in helical CT are generally undersampled by a factor of at least 2 in the longitudinal direction and because the exploitation of the redundancy of fanbeam data acquired over 360° to generate additional longitudinal samples does not automatically eliminate the aliasing. In this paper we demonstrate that for pitches near 1 or lower, the redundant fanbeam data, when used properly, can provide sufficient information to satisfy a generalized sampling theorem and thus to eliminate aliasing. We develop and evaluate a Fourier-based algorithm, called 180FT, that accomplishes this. As background we present a second Fourier-based approach, called 360FT, that makes use only of the directly measured data. Both Fourier-based approaches exploit the fast Fourier transform and the Fourier shift theorem to generate from the helical projection data a set of fanbeam sinograms corresponding to equispaced transverse slices. Slice-by-slice reconstruction is then performed by use of two-dimensional fanbeam algorithms. The proposed approaches are compared to their counterparts based on the use of linear interpolation-the 360LI and 180LI approaches. The aliasing suppression property of the 180FT approach is a clear advantage of the approach and represents a step toward the desirable goal of achieving uniform longitudinal resolution properties in reconstructed helical CT volumes.