Improvement in C-Arm Acquired DSA Image Quality via Combined Effect of Inverse Consistent Motion Correction and Nonlinear Normalization

In this paper, we present a new two-step strategy to improve DSA image quality. First, an inverse consistent image registration algorithm is used to register the mask frame and the bolus frame before subtraction. This can effectively reduce the motion artifacts in DSA. Second, the resulted DSA image is further enhanced by background diffusion and nonlinear normalization for better visualization. The proposed method has been evaluated on a database of 73 subjects by quantitatively measuring signal-to-noise (SNR) ratio and Tenengrad measure. DSA embedded with proposed strategies demonstrates an improvement of 136.56% over conventional DSA in terms of SNR. Our system runs on Eigen´s DSA workstation using C++ in Windows environment.