Region-of-Interest CT Reconstruction using Object Extent and Singular Value Decomposition

In computed tomography, a whole scan of the object may be impossible, generally because is larger than scanner field view. Such set up leads to truncated projections. Using differentiated backprojection, reconstruction problem can reduced 1-D problems consisting inversion Hilbert transform. When partly overlaps view, this commonly referred as “one-sided transform.” Our work investigates situation and proposes novel approach address it. extent supposedly known a priori , pseudoinverse transform by singular value decomposition, its values are replaced simple estimation. The estimation calculated using decomposition convex hull filled with constant per line from corresponding projection in direction experiments illustrate image quality improvements resulting compared truncation speed improvement 2-D iterative solving penalized least squares conjugate gradient algorithm.