Dedicated mammographic units should be equipped with anti-scatter grids (ACR, 1993; AHCPR, 1994). Scattered radiation can cause a significant reduction in subject contrast in mammogra-phy resulting in impaired detection of calcifications and the outlines of tumor masses. The advent of specialized mammographic grids revolutionized the radiologist's ability to evaluate dense tissue (Barnes and Brezovich, 1978; Chan et al., 1985; Dershaw et al, 1985; Egan et al., 1983; Friedrich and Weskamp, 1978; Jost, 1979; Logan and Stanton, 1979; Sickles and Weber, 1986; Stanton and Logan, 1979).

The grid (Figure 3.11) placed between the breast and the image receptor, absorbs scattered radiation that would otherwise reach the image receptor, improves contrast, and results in better definition of

Fig. 3.11. Mammography bucky assembly. Black lines in the grid represent radiopaque lead strips that make up the grid. The lead strips are focused to the focal spot. Arrow indicates that the grid moves through a distance of >20 grid line spacings, that is, >20 (d + D) (where, d = width of lead lamellae and D = width of radiolucent interspace material) (Barnes, 1999).

Fig. 3.11. Mammography bucky assembly. Black lines in the grid represent radiopaque lead strips that make up the grid. The lead strips are focused to the focal spot. Arrow indicates that the grid moves through a distance of >20 grid line spacings, that is, >20 (d + D) (where, d = width of lead lamellae and D = width of radiolucent interspace material) (Barnes, 1999).

the borders of glandular tissues. However, even with its advantages, the use of a grid does not eliminate the need for firm compression to spread apart the glandular tissues and to permit better visualization of the borders of small lesions. The use of a grid does result in increased patient dose and exposure time. However, units with high output can maintain exposure times at levels that do not create significant patient motion and film reciprocity law failure problems (Villafana, 1990).

The intensity of scattered radiation (S) reaching the image receptor [relative to the primary radiation intensity (P) at the same point] is described by S/P. In mammography, if a grid is not used, S/P can vary from 0.33 to 1 as the diameter of the radiation field increases from 4 to 14 cm and the breast phantom thickness increases from 3 to 6 cm (Barnes and Brezovich, 1978). Even higher S/Ps are associated with thicker breasts (e.g., S/P = 1.5 at a thickness of 8 cm). The effect of such scattered radiation is to reduce contrast and the magnitude of the effect is described by the scatter degradation factor (SDF) where SDF = 1/[1 + (S/P)] (Barnes, 1994). Thus, at an S/P of 0.33, only 75 percent of the available contrast will be imaged and at a S/P of one, only 50 percent of the contrast will be imaged. Control of scattered radiation, therefore, has the potential for significantly improving contrast (Figure 3.12).

Scattered radiation can be reduced by a factor of three through the use of an appropriate grid (AAPM, 1990; Yaffe, 1991). Since the grid absorbs 50 percent or more of the radiation beam, the contrast improvement is achieved only at the expense of increasing the exposure by a factor of 2 to 2.5 compared with nongrid techniques (AAPM, 1990; NCRP, 1986). It is possible to offset at least some of this increased exposure by increasing the operating potential (Friedrich and Weskamp, 1978).

Grids specifically designed for mammography are necessary since the materials and construction of general radiographic grids result in excessive attenuation of the unscattered portion of the low-energy mammographic x-ray beam, as well as increased geometric unsharpness due to the thickness of the grid assembly (ACR, 1993; Feig, 1987; Friedrich and Weskamp, 1978; NCRP, 1986). Special purpose mammographic grids are extremely thin, with lead grid strips, or septa, only about 1 mm in height. The septa are typically 16 pm thick and the interspaces are about 300 pm wide (Feig, 1987). The grid ratio (the height of septa relative to the distance between the septa) is usually in the range of 4:1 to 5:1 and the grid should have about 32 septa (or "lines") per centimeter. To minimize attenuation of the primary (image forming) radiation and

Nurse Patient Ratio Grid

Breast Thickness (cm)

Breast Thickness (cm)

Fig. 3.12. Plots of the contrast improvement factor (CIF) and bucky factor (BF) of a typical mammography grid versus breast thickness (Barnes, 1999).

avoid an unnecessary increase in patient dose, the interspace and the grid cover materials should have a low x-ray attenuation and have a radiographically uniform structure (ACR, 1993; Yaffe, 1991). For this reason, the interspace material is usually fiber (paper) and the grid cover is often carbon fiber. The use of such materials will also reduce the extent to which the patient dose must be increased to compensate for the absorption by the grid. The typical grid ratio and "bucky factor" (the ratio of the milliampere seconds required with the grid to that required without the grid to obtain a given film optical density at a typical clinical operating potential) should be indicated on a label on the grid, as well as on the outside of the grid assembly. Disassembly of the equipment should not be required to verify the specifications of the components.

Recently, rhombic cellular structure air interspaced grids have been introduced (Figure 3.13). These grids have the potential to improve image contrast (and reduce grid absorption) compared with conventional grids (Figure 3.14).

Moving grids are widely used and a mammographic unit should be equipped with a mechanism designed to move the grid during the x-ray exposure, in such a manner, that the grid septa are not visible on the mammographic image. Images of a uniform phantom

High Transmission Cellular Grids
Fig. 3.13. High transmission cellular grid multidirectional scattered radiation absorption (courtesy of Lorad Corporation, Danbury, Connecticut) (Haus, 1999b).

taken after the grid system is activated and with only the compression device in the x-ray beam should not reveal any grid lines or other density fluctuations, since these will degrade the image. Moving grids may produce grid lines on mammograms when exposures are long enough to cover several oscillations of the grid. In such a case, the grid lines may be strongly imaged during those brief periods when the septa are at rest as the motion of the grid is reversed. Grid line artifacts may also occur at the end of very long exposures, if the grid oscillations diminish gradually (Dance et al., 1992). Such artifacts may also be produced when exposures are very brief or the oscillations are too slow. In this case, not enough time elapses for the images of the grid lines to be "averaged out." This type of problem can be more significant when high-speed screen-film combinations are used. In order to avoid these problems, it should be insured that test images of the moving grid show no grid lines or artifacts over a range of uniform phantom thicknesses from 2 to 6 cm at optical densities of about 1.3 (ACR, 1993).

HTC vs Li near-

Typical % Improvement

Was this article helpful?

0 0

Post a comment