Evaluation Of Spatial Performance For Cone-Beam Computed Tomography

X-ray computed tomography(XCT)is an important method in medical diagnosis and other fields.Cone-beam XCT imaging,due to the low radiation dose,high utilization of radiation,high scanning speed and other advantages,has gained rapid development with the maturity of flat-panel detector.Imaging quality and radiation dose are two focuses of XCT imaging research at present.The image quality is influenced by the configuration of the system parameters,while in practice,it needs constant trial and optimization,which consumes a lot of time as well as human cost.Monte Carlo simulation can economically and effectively compare different parameters of the imaging system to save resources and improve efficiency.A cone-beam XCT Monte Carlo simulation system is built to employ the multi-energy X-ray source,giving full consideration to the imaging process including the microscopic effect,to achieve accurate simulation of the actual imaging process with different imaging conditions.To reduce the radiation dose under the condition of guaranteeing the image quality is a key problem in XCT imaging field,and the developing low-dose XCT reconstruction method is one of the effective solutions.At present,some scholars have developed the low-dose cone-beam XCT reconstruction algorithms based on the compressive sensing theory,but their performance has not been fully analyzed,ending up with having not yet been applied in practical instruments.In order to promote the research of low-dose reconstruction,we use the cone-beam XCT Monte Carlo simulation system to obtain the projections of the different configurations,which are to be reconstructed by compressive sensing algorithms.After obtaining corresponding reconstruction results,we then compare the results using noise equivalent quanta(NEQ),a kind of comprehensive evaluation method combining the spatial resolution and noise level of the system.In the NEQ evaluation mode,the system spatial resolution is evaluated by Modulation Transfer Function(MTF),and the noise level is evaluated by Noise Power Spectrum(NPS).Based on the accurate NEQ evaluation method,the feasibility and practicability of the compressive sensing algorithms are evaluated by comparing the results of both the compressive sensing reconstruction algorithms and the conventional cone-beam reconstruction algorithm based on filtered back-projection.