Anam Choirul, Fujibuchi Toshioh, Budi Wahyu Setia, Haryanto Freddy, Dougherty Geoff
Department of Physics, Faculty of Mathematics and Natural Sciences, Diponegoro University, Semarang, Central Java, Indonesia.
Department of Health Sciences, Faculty of Medical Sciences, Kyushu University, Fukuoka, Fukuoka Prefecture, Japan.
J Appl Clin Med Phys. 2018 Nov;19(6):244-252. doi: 10.1002/acm2.12476. Epub 2018 Oct 19.
The purpose of this study was to introduce a new algorithm for automated measurement of the modulation transfer function (MTF) using an edge of a readily available phantom and to evaluate the effect of reconstruction filter and field of view (FOV) on the spatial resolution in the CT images.
Our automated MTF measurement consisted of several steps. The center of the image was established and an appropriate region of interest (ROI) designated. The edge spread function (ESF) was determined, and a suitably interpolated ESF curve was differentiated to obtain the line spread function (LSF). The LSF was Fourier transformed to obtain the MTF. All these steps were accomplished automatically without user intervention. The results of the automated MTF from the edge phantom were validated by comparing them with a point image, and the results of the automated calculation were validated by the standard fitting method. The automated MTF calculation was then applied to the images of two polymethyl methacrylate (PMMA) phantoms and a wire phantom which had been scanned by a Toshiba Alexion 4-slice CT scanner and reconstructed with various filter types and FOVs.
The difference in the 50% MTF values obtained from the edge and point phantoms were within ±4%. The values from the automated and fitted methods agreed to within ±2%, indicating that the automated MTF calculation was accurate. The automated MTF calculation was able to differentiate MTF curves for various filters. The spatial resolution values were 0.37 ± 0.00, 0.71 ± 0.01, and 0.78 ± 0.01 cycles/mm for FC13, FC30 and FC52 filters, respectively. The spatial resolution of the images decrease linearly (R > 0.98) with increasing FOVs.
An automated MTF method was successfully developed using an edge phantom, the PMMA phantom. The method is easy to implement in a clinical environment and is not influenced by user experience.
本研究的目的是引入一种新算法，用于使用现成模体的边缘自动测量调制传递函数（MTF），并评估重建滤波器和视野（FOV）对CT图像空间分辨率的影响。
我们的自动MTF测量包括几个步骤。确定图像中心并指定合适的感兴趣区域（ROI）。确定边缘扩展函数（ESF），并对适当插值的ESF曲线进行微分以获得线扩展函数（LSF）。对LSF进行傅里叶变换以获得MTF。所有这些步骤均无需用户干预即可自动完成。通过将边缘模体的自动MTF结果与点图像进行比较来验证其结果，并通过标准拟合方法验证自动计算的结果。然后将自动MTF计算应用于两个聚甲基丙烯酸甲酯（PMMA）模体和一个线模体的图像，这些模体已由东芝Alexion 4层CT扫描仪扫描，并使用各种滤波器类型和FOV进行重建。
从边缘模体和点模体获得的50%MTF值差异在±4%以内。自动方法和拟合方法得到的值在±2%以内一致，表明自动MTF计算是准确的。自动MTF计算能够区分各种滤波器的MTF曲线。对于FC13、FC30和FC52滤波器，空间分辨率值分别为0.37±0.00、0.71±0.01和0.78±0.01周期/毫米。图像的空间分辨率随着FOV的增加而线性下降（R>0.98）。
使用边缘模体PMMA成功开发了一种自动MTF方法。该方法易于在临床环境中实施，且不受用户经验的影响。
