Aiming at the problem of inaccurate imaging model of three-dimensional (3D) reconstruction of rotational digital subtraction angiography (DSA), a nonlinear model based on object-oriented quantization is introduced first. The model quantizes the projection pixel of 3D vessel slice as the number of vessels which the X-ray goes through. Furthermore, a slice reconstruction algorithm named segment self-guide reconstruction (SSGR) is developed. The SSGR is especially suitable for solving the problem of sparse projection and limited-views. Finally, the simulated results have proved the feasibility of the model and the validity of the algorithm.
Algorithms ; Angiography, Digital Subtraction ; Computer Simulation ; Image Processing, Computer-Assisted ; methods ; Imaging, Three-Dimensional ; Models, Cardiovascular ; Nonlinear Dynamics

This article reviews the research progress in qualitative evaluation of medical image reconstruction, especially the binary image reconstruction. According to the affected point and evaluation character, the evaluation criteria are classified into four categories. Based on the brief introduction of all kinds of evaluation criteria, the paper makes an analytical comparison, and points out their merit and fault. This study provides a useful basis for selection and design of medical image reconstruction algorithms.
Algorithms ; Evaluation Studies as Topic ; Image Processing, Computer-Assisted ; methods ; standards