Image reconstruction from a reduced number of projections in Micro-CT specimen imaging

摘要

Micro-CT is a useful imaging tool for a wide range of biomedical research and preclinical applications. Accurate three-dimensional images of the subject such as vasculature specimens can be obtained and further be used to derive quantitative anatomic and physiologic information. In this work, we applied the adaptive-steepest-descent-projection-onto-convex-sets (ASD-POCS) algorithm to reconstructing micro-CT images from real data of animal vasculature specimens acquired at reduced number of views. We carried out evaluation studies on sparse-view reconstruction quality, and results show that images with quality comparable to that achieved in standard micro-CT specimen imaging can be obtained with the ASD-POCS algorithm from data much less than that required by current micro-CT scans.