An RIP-based evaluation method for candidate next generation cardiac CT architectures with carbon nanotube x-ray source

摘要

Despite impressive advancements in computed tomography (CT) technology in recent years, there are still critical andimmediate needs in cardiac CT in terms of high spatial resolution, high temporal resolution and low radiation dose.Because carbon nanotube (CNT) x-ray sources can be compactly integrated, this technology can be used for multisource or stationary system to improve temporal resolution. To avoid the rotation of x-ray source, a lot of source-detector pairs are needed in a stationary CNT-based x-ray system. Limited by the space and costs, the number of source-detector pairs could not be too large, which results in a few-view scan problem. The reconstruction can be modeled as a lsub1/sub-norm minimization problem, which usually can be solved by compressive sensing (CS) based algorithms. To evaluate the data completeness of candidate next generation cardiac CT architectures with CNT x-ray source, based on the fact that smaller restricted isometry property (RIP) constants lead to a better lsub1/sub-norm recovery, we construct a measurement related to the RIP constants. The results show that the proposed RIP-based evaluation method coincides with the known CT reconstruction theory. This method is simple and easy to be implemented for different CT scan architectures, and it provides a practical tool to evaluate the data completeness in the framework of lsub1/sub-norm recovery theory without a specific CS-based reconstruction algorithm.© (2012) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.