A residual correction method for iterative reconstruction with inaccurate system model

The quality of images produced by iterative image reconstruction methods is directly affected by the accuracy of the system model being used. Although the system response of an imaging system can be computed or measured with high accuracy, practical constraints on computation cost often force the adoption of various approximations to the system model to obtain a computationally efficient system matrix. These approximations inevitably cause artifacts in reconstructed image. In this work we propose a residual correction method for iterative reconstruction to reduce reconstruction artifact caused by the model inaccuracies. Unlike conventional iterative methods which assume that the system matrix is accurate, the proposed method reconstructs an initial image with an approximate system model and then corrects for the reconstruction artifacts by introducing a data correction term to compensate for the model inaccuracies. Computer simulation showed that the proposed method can significantly improve image quality in terms of objective function value compared to approximate algorithms, while it is computationally more efficient than the conventional method that uses the accurate system model at every iteration.