【摘要】 目的:在锥形束CT(CBCT)扫描中设计一种高信噪比去除锥角伪影方法,以提高螺旋大锥角宽检测器的图像质量。方法:从数据使用角度分析螺旋锥角伪影产生的原因,通过对产生螺旋锥角伪影的数据进行不同的加权处理,使用加权滤波反投影算法,分别得到高信噪比有锥角伪影图像和低信噪比无锥角伪影图像,针对低信噪比图像结合高信噪比图像使用引导滤波的方式实现高信噪比锥角伪影的去除。结果:通过使用临床数据对加权滤波反投影算法与原始算法进行对比,图像中的锥角伪影得到了有效抑制,同时保持了原有重建方法的信噪比。结论:高信噪比去除锥角伪影方法在大锥角重建时能够在保证高信噪比图像时有效抑制锥角伪影,提高螺旋大锥角宽检测器的图像质量。更多还原

【Abstract】 Objective:To design a high signal-to-noise ratio(SNR) method that could remove cone angle artifacts in spiral scanning so as to improve the image quality for spirally large cone angle detectors.Methods:This article analyzed the causes of spiral cone angle artifacts from using data.Through implemented different weighted approach for these data that produced spiral cone angle artifacts,and used weighted filtering back-projection algorithms to obtain high SNR images with cone angle artifact and low SNR images without cone angle artifacts.The low SNR images combined with high SNR images to use guided filtering mode so as to realize the remove for cone angle artifacts of high SNR images.Results:Through used clinical data to compare weighted filtering back-projection algorithms and original algorithms.And the cone angle artifacts in the image were effectively suppressed.At the same time,the SNR of former reconstruction method was maintained.Conclusion:The high SNR method for removing cone angle artifacts can effectively suppress cone angle artifacts and improve image quality of spiral detector with large cone angle when ensures high SNR images.更多还原