Regularization for uniform spatial resolution properties in penalized-likelihood image reconstruction

Traditional space-invariant regularization methods in tomographic image rec onstruction using penalized-likelihood estimators produce images with nonun iform spatial resolution properties. The local point spread functions that quantify the smoothing properties of such estimators are space-variant, asy mmetric, and object dependent even for space-invariant imaging systems. We propose a new quadratic regularization scheme for tomographic imaging syste ms that yields increased spatial uniformity and is motivated by the least-s quares fitting of a parameterized local impulse response to a desired globa l response. We have developed computationally efficient methods for PET sys tems with shift-invariant geometric responses. We demonstrate the increased spatial uniformity of this new method versus conventional quadratic regula rization schemes in simulated PET thorax scans.