Study of Monte Carlo Simulator for Estimation of Anti-Scatter Grid Physical Characteristics on IEC 60627:2013-Based

The anti-scatter grid (grid) is a major component of X-ray imaging devices, improving the quality of the image by removing X-rays scattered while passing through the subject. However, problems such as image distortions or increasing the dose of a patient unnecessarily may result if the grid is not suitable for a specific digital x-ray receptor (detector.) Selecting a suitable grid can take significant work and testing due to the wide range of specifications and physical characteristic of both grid and digital imaging x-ray detectors. In order to reduce the time cost and to improve the accuracy of selecting a suitable grid, this study implemented a Monte Carlo simulation for estimating the physical characteristics of the grid and verified the accuracy of the result by comparing with the physical characteristics of the actual grid. For the verification, this study compared the estimated physical characteristics with the measured physical characteristics for ten (10) grids with different specifications. The physical characteristics were measured at RQR (Radiation Qualities in Radiation Beams emerging from the X-ray Source Assembly) 4/6/8/9 of the Radiation conditions and analyzed Transmission of Primary radiation (Tp), Transmission of Total radiation (Tt) and Transmission of Scattered radiation (Ts) of the physical characteristics of the grid. As a result of the analysis, less than 1% average deviation between simulation and physical measurement was observed with all ten (10) grids. The changes of the physical characteristics as the specifications (line density and ratio) of the grid changed were also evaluated, and found to have a Pearson’s correlation coefficient of 0.998 between simulation and measurements. From the above results, the proposed program in this paper is judged reasonable as a grid physical characteristics prediction program.


Introduction
X-ray is one of the greatest discoveries of the past century, and x-ray scanners are used in various fields, including medical, non-destructive testing, and security. The antiscattered grid was first introduced in 1913 and was used to improve image quality by removing scattering lines created during X-ray imaging process. The scattered ray refers to the x-ray that loses its straightness and gets refracted when passing through the subject. If this scattered ray reaches the detector, it results in an image quality degradation by making the image blurry. The grid is placed between the image receptor and the subject, improving the quality of X-ray images through removal of scattered rays and thus contributing to accuracy of diagnosis. However, using an unoptimized grid per X-ray imaging receptor may create problems such as image distortion, increased dose of x-ray for patient, etc. To prevent this, it is critical to use the grid which is optimized per X-ray imaging receptor used. The Xray imaging device and the grid have a wide range of specifications and characteristics, requiring a great deal of time and effort for the selection of optimized grid for X-ray imaging device.
This study developed simulations to predict the physical characteristics of the grid based on Monte Carlo method and verified the accuracy of the simulation by comparing to physical characteristics of the grid.
Chapter 2 introduces a grid used for simulation and chapter 3 describes the characteristic values for extraction. In chapter 4 describe how to extract the characteristic values and the device composition and demonstrates the performance in accordance with the experiment in chapter 5.

Figure 1.
Acquisition of x-ray image with grid in DR system.

The Composition and Specifications of Grid
Grid consists of pairs of interspacer strip and absorber strip with width of 60~400µm and 15~50µm respectively. The lowdensity material such as aluminum, etc. is used for interspacer and it lets primary radiation pass through to reach the detector. High-density material such as lead is used for absorber and it absorbs and removes the scattered rays that are created when x-ray passes through the subject. Line density refers to the number of pairs of inter-spacer and absorber strip in a given unit distance and Line-Pairs/CM (LP/CM) or Line-Pairs/Inch (LP/IN) is used as the unit. The line density is an important factor as it determines the occurrence of Moiré phenomenon in image and physical characteristics of grid. Grid ratio stands for the ratio of the grid height against the width of interspacer and it is also one of the main factors that determines the grid physical characteristics. In the grid, each pair of interspacer and absorber stand in the direction of the x-ray source at a specific angle based on straightness of the x-ray from the source. This angle alignment is determined with consideration to the distance from the bottom surface of grid to the x-ray source, referred to as FD (Focal Distance). Thin aluminum or carbon fiber sheets are applied to both sides of grid surfaces to protect the grids. Type of material and the thickness used for the cover also has impact on grid's physical characteristics.

Physical Characteristics of Gird
In general, There are 7 physical characteristics of the proper grid required for the X-ray imaging, and it must follow IEC (International Electrotechnical Commission) 60627: 2013 (Diagnostic X-ray imaging equipment -Characteristics of general purpose and mammographic antiscatter grids) that is enacted by International Electrotechnical Commission (IEC) [6]. Three of the physical characteristics are obtained from the actual measurement, which are Tp (Transmission of Primary radiation), Tt (Transmission of Total radiation), and Ts (Transmission of Scattered radiation.) B (Grid Exposure Factor), K (Contrast Improvement Ratio), Σ (Selectivity), and Q (Image Improvement Factor) are four physical characteristics, which are obtained by calculating measured values, Tp refers to the amount of primary x-ray radiation from the x-ray tube that actually reaches the detector after passing through grid. Higher value for Tp means better performance.
Tp can be calculated according to the following formula [7].

=
(1) P g : Primary radiation with grid P ng : Primary radiation without grid Ts refers to the amount of the scattered x-ray radiation generated while the x-ray passing through the object that actually reaches the detector even after going through the grid. Lower value for Ts means better performance. Ts can be calculated according to the following formula.

=
(2) S g : Scatter radiation with grid S ng : Scatter radiation without grid Tt refers to the amount of both primary and scattered xrays that reach the detector after passing through the grid. Higher value for Tt means better performance. Tt can be calculated according to the following formula.

=
(3) T g : Total radiation with grid T ng : Total radiation without grid B stands for the degree indicator of the radiation dose that is increased to compensate for X-rays which is reduced by the Xray grid and the lower B means the better the performance. B can be calculated according to the following formula.

= (4)
K stands for the effect indicator of the contrast which is improved by using the X-ray grid and the higher K means the better the performance. K can be calculated according to the following formula.
Σ stands for the proportion indicator of the transmitted primary X-ray by comparing with the scattered X-ray when using the grid and the higher Σ means the better the performance. Σ can be calculated according to the following formula. = Q stands for showing how much the SNR (Signal to Noise Ratio) of detector increases when using the grid and the higher Q means the better the performance. Q is calculated according to the following formula.

1) IEC Test Fixture
In this study, the IEC Test Fixture, in Figure 6, is used to measure three physical characteristics of grid's, Tp, Tt, and Ts. IEC Test Fixture is a device based on IEC 60627: 2013 standards. The Toshiba's E7242X X-ray tube with tube voltage range of 40~125kV, maximum current of 290 to 800 mA, and a focal spot of 0.6 and 1.5 mm is used for this test fixture. The Toreck's EY-1300 is used as radiation dosimeter and the scintillator is Gd2O2S: Tb with an area density of 68 mg / cm 2 , and the PMT (Photomultiplier tube) is used as a photo-detector.  Table 1 shows the list of ten grids used to compare the results of simulation data against the actual physical characteristics.

3) Condition of the radiation
The condition of the radiation is measured by the three conditions of RQR 4, RQR 6 and RQR 9, which are the basic requirements of IEC 60627: 2013, and the additional requirement of the customer, RQR 8. Each condition of the radiation's filter (thickness, material) are set as shown in Table 2 based on IEC 61267: 2005 [9].  Figure 7 shows that the user interface of the program used to simulate and estimate characteristics of the grid. The program implements grid and geometry measurement of grid for physical characteristics using MCNPX as well as the GUI of program in Visual Basic language. The program receives the specification of grid (line density, absorber, inter-spacer, ratio, FD and cover type) and condition of the radiation as inputs then displays the estimated performance on the output screen from the simulation. The frequency of simulation (nps) is set as 8 ~ 14 X 10 8 in order to increase the reliability of the estimated grid performance.

5) Measurement of grid physical characteristics
The physical characteristic of the grid is measured using the IEC Test Fixture in Figure 6. The physical characteristic is measured by measuring the radiation dose in each physical characteristics measurement structure with/without the grid, then calculated the measured radiation dose according to each formula of physical characteristics. In order to improve the reliability of the radiation dose measurement, we used the average radiation dose as representative for each grid which Max/Min value is excluded from 8 times measurement out of 10 times.  Table 3 shows that the measured and the estimated value of physical characteristics for the grid's Tp, Tt, and Ts when the radiation condition was set to RQR 4. In RQR 4, the mean error between the measured and the estimated value of Tp was 1.04% while the maximum error was 1.96%. The mean error between the measured and the estimated value of Tt was 0.83% while the maximum error was 1.99%. The mean error between the measured and the estimated value of Tt was 0.83% while the maximum error was 1.81%. In RQR 4, the mean error between the measured and the estimated value of all grids was 0.90% while the maximum error was 1.99%.

1) Comparison for Characteristic of gird in RQR 4
2) Comparison for Characteristic of grid in RQR 6 Table 4 shows that the measured and the estimated value of physical characteristic for the grid's Tp, Tt, and Ts when the condition of radiation was set to RQR 6. In RQR 6, the mean error between the measured and the estimated value of Tp was 0.81% while the maximum error was 1.80%. The mean error between the measured and the estimated value of Tt was 0.86% while the maximum error was 1.96%. The mean error between the measured and the estimated value of Tt was 0.88% while the maximum error was 1.82%. In RQR 6, the mean error between the measured and the estimated value of all grids was 0.85% while the maximum error was 1.96%. 3) Comparison for Characteristic of grid in RQR 8 Table 5 shows that the measured and the estimated value of physical characteristics for the grid's Tp, Tt, and Ts when the condition of radiation was set to RQR 8. In RQR 8, the mean error between the measured and the estimated value of Tp was 1.25% while the maximum error was 1.95%. The mean error between the measured and the estimated value of Tt was 1.18% while the maximum error was 1.94%. The mean error between the measured and the estimated value of Tt was 0.96% while the maximum error was 1.87%. In RQR 8, the mean error between the measured and the estimated value of all grids was 1.13% while the maximum error was 1.95%.  Table 6 shows that the measured and the estimated value of physical characteristics for the grid's Tp, Tt, and Ts when the radiation condition was set to RQR 9. In RQR 9, the mean error between the measured and the estimated value of Tp was 1.07% while the maximum erro was 1.96%. The mean error between the measured and the estimated value of Tt was 1.04% while the maximum error was 1.94%. The mean error between the measured and the estimated value of Tt was 0.96% while the maximum error was 1.87%. In RQR 8, the mean error between the measured and the estimated value of all grids was 1.13%, and the maximum error was 1.95%.   Figure 8 shows a graph of the change in physical characteristics according to grid ratio and the grid has same specification (Line Density 103 LP/in, FD 100 cm, Al cover) except grid ratio. In RQR 6, the measured and the estimated value of Tp are decreased by 8.6% and 7.6%, respectively as the ratio increases from 6:1 to 12:1. In the same condition, the measured and the estimated value of Tt are decreased by 15.0% and 14.2%, respectively. Also, the measured and the estimated value of Ts are decreased by 16.0% and 15.8%, respectively.

5) Changes of characteristics according to the ratio
In RQR 8, the measured and the estimated value of Tp are decreased by 7.8% and 7.1%, respectively as the ratio increases from 6:1 to 12:1. In the same condition, the measured and the estimated value of Tt are decreased by 14.3% and 16.7%, respectively. Also, the measured and the estimated value of Ts are decreased by 16.6% and 17.6%, respectively.
As the grid ratio increases, the physical characteristic Tp, Tt, and Ts of measured and estimated physical characteristic was shown as Pearson correlation coefficient 0.998. [10]. Therefore, it can be considered as measured and estimated physical characteristics show a similar tendency of change.
6) Changes of characteristics according to the line density Figure 9 shows a graph of the change in physical characteristics according to the line density and the grid has same specification (Grid Ratio 8:1, FD 100 cm, Al cover) except line density. In RQR 6, the measured and the estimated value of Tp are increased by 5.8% and 4.5%, respectively as the ratio increases from 6:1 to 12:1. In the same condition, the measured and the estimated value of Tt are increased by 7.8% and 7.3%, respectively. Also, the measured and the estimated value of Ts are increased by 8.8% and 9.8%, respectively.
In RQR 8, the measured and the estimated value of Tp are increased by 4.1% and 1.8%, respectively as the ratio increases from 6:1 to 12:1. In the same condition, the measured and the estimated value of Tt are decreased by 8.6% and 9.1%, respectively. Also, the measured and the estimated value of Ts are decreased by 11.9% and 11.9%, respectively.
As the line density changes, the physical characteristic Tp, Tt, and Ts of measured and estimated physical characteristic was shown as Pearson correlation coefficient 0.998. Therefore, it can be considered as measured and estimated physical characteristics show a similar tendency of change.

Conclusion
This study developed a program for estimating a grid's physical characteristics based on IEC 60627: 2013, and compared the physical characteristics of measured grids with the estimates. The difference between the measured and the estimated physical characteristics averaged 0.90% (maximum 1.99%) in RQR 4, 0.85% (maximum 1.96%) in RQR 6, 1.13% (maximum 1.95%) in RQR 8 and 1.04% (maximum 1.96%) in RQR 9. The average difference in all conditions was 0.98%, less than 1%. Also, changes in the grid ratio and line density produced very similar variation in the measured and simulated physical characteristics of the grids.