Compressed Breast Thickness

Compressed Breast Thickness

Fig. 3.14. Contrast improvement factor versus compressed breast thickness for high-transmission cellular (HTC) and linear grid. This is a relative curve. It changes with different breast composition and film gradient (courtesy of Lorad Corporation, Danbury, Connecticut) (Haus, 1999b).

With a moving grid, the grid assembly should be sufficiently rigid so that the grid motion is not impeded when the breast being imaged is under firm compression. This degree of rigidity can be demonstrated by placing a 4 cm thick (approximate) cassette sized phantom made of either acrylic or BR-12®,6 0.5 cm thick, in the center of the imaging area and compressing it using the full pressure of the compression device (ACR, 1999). Under these conditions, test images should demonstrate that the motion of the grid is not impeded. The grid cover should be of uniform construction so that structural artifacts are not superimposed on the mammographic image. As mentioned above, the grid itself should be uniform and have no regions of increased attenuation that would produce image artifacts.

Two sizes of moving grids are necessary: (1) for small breasts to accommodate cassettes for 18 x 24 cm film, and (2) for large breasts to accommodate cassettes for 24 x 30 cm film. Although many

6BR-12® is an epoxy resin-based tissue substitute (Gammex, Middle-ton, Wisconsin) (White et al., 1977).

patient's breasts can be accommodated on an 18 x 24 cm film, approximately 20 percent of patients require a 24 x 30 cm film to include the axillary tail (ACR, 1993). One large grid equipped with devices to hold both the smaller and the larger cassettes will not suffice. When a small breasted woman lifts her arm above the larger tray, the skin becomes taut, which could prevent the technologist from pulling the patient's breast forward on the film and could result in missing a posterior cancer.

The design of the unit's C-arm should be such that it facilitates switching from one grid size to another, as well as removing the grid entirely in those few cases where use of a grid would be inappropriate. Ideally, the mammographic unit should be equipped with an interlock feature that will prevent exposures when the grid is not in place or is in place but is disconnected from the unit, unless special action is taken to override the interlock. Such a feature would prevent the technologist from inadvertently making an exposure with the grid disconnected.

Despite the advantages of grids and the significant additional clinical information their use provides, these devices do have certain disadvantages. As noted above, the grid will absorb >50 percent of the radiation leaving the breast and compensating for this reduced exposure rate requires doubling the patient's dose. In general, the greater the contrast improvement provided by the grid the more the dose will need to be increased. Higher operating potential settings, increased filtration, increased exposure time, or use of a higher speed screen-film combination (or some combination of these factors) can counteract the higher dose, but not without corresponding consequences. The harder beams associated with higher operating potentials or greater filtration reduce subject contrast, undermining the grid's effectiveness. Longer exposure times can result in patient motion problems and may necessitate even greater increases in patient dose due to film reciprocity law failure. Faster imaging systems can result in significant increases in quantum noise and a consequent reduction in image quality.

Despite its minor disadvantages, the grid is absolutely essential for assessing dense glandular tissue and has revolutionized modern mammography. Previously, many radiologists only used the grid for patients with dense glandular tissue or breast tissue that could not be compressed to <6 cm (Dershaw, 1987; NCRP, 1986; Sickles and Weber, 1986). Because of the dose reductions achievable with newer screen-film combinations, the greater contrast enhancement capability of modern grids and the availability of higher output units (with shorter exposure times), using a grid for virtually every routine mammogram has become common practice (Feig, 1987; NCI, 1993). See Table 3.13 for desirable characteristics of grids.

3.1.10 Magnification Mammography

Dedicated mammographic units intended for diagnostic or problem-solving mammography must have the capability of doing magnification mammography (AHCPR, 1994). A small focal spot should be provided for imaging in the magnification mode (ACR, 1993), as well as a magnification stand designed to support the breast in an elevated position, significantly above the plane of the image receptor. When imaged in this elevated position, a geometrically magnified image of the breast is produced. Such magnification images can often provide clinically significant information concerning microcalcifications and the borders of masses that cannot be obtained from nonmagnification images and are, therefore, useful in distinguishing malignant from benign breast disease (Sickles, 1979; 1980; 1987a; Sickles et al, 1977). In magnification mammography, image quality is improved for a variety of reasons

Table 3.13—Desirable characteristics of grids.

• Grid properties

- Extremely thin septa of about 1 mm in height

- Septa typically 16 pm thick

- Septa interspace is about 300 pm wide

- Ratio usually 4:1 or 5:1 with about 32 septa (lines) per centimeter

- Septa are radiographically uniform structures

- Interspace material between septa is usually fiber (paper)

- Cover is made of carbon fiber for low x-ray attenuation

- Ratio and bucky factor is indicated on label on grid, as well as on outside of grid assembly

- No visible grid lines for AEC exposures of phantom thicknesses from 2 to 6 cm

• Bucky properties

- Cover rigid enough so grid motion is not impeded under compression

- Two sizes of grids for the two film sizes of 18 x 24 cm and 24 x 30 cm

- Allows for easy switch of grids

- Interlock to prevent exposure when grid is not in place or in place but disconnected from unit

(Haus, 1990; Haus et al., 1979). Magnification increases the effective resolution of the image receptor, because every detail in the enlarged image of the breast is magnified. Small structures, whose visibility was limited by the image-receptor blur during contact mammography, are imaged at a magnified size where the effect of blur is reduced. In addition, there is a decrease in the effective noise since quantum noise is not magnified. There is also a decrease in the scattered radiation that is recorded in the image due to the introduction of an air gap between the exit surface of the breast and the image receptor (Barnes, 1979; Barnes and Brezovich, 1978; Nielson and Fagerberger, 1986). The typical air gap is 16 to 30 cm for a 1.5 to 2 magnification factor depending on the SID. The increased dose associated with magnification mammography, as well as potential film reciprocity law failure problems can be offset by the use of a faster image receptor (Bassett et al., 1981).

The resolution improvement that can be achieved by magnification mammography depends on the size of the focal spot (and the location of the structure of interest in the breast) since image resolution is ultimately limited by geometric unsharpness (Haus et al., 1979). The degree of magnification provided should be between 1.5 and 2 times depending on the actual size of the focal spot, the unit geometry, SID, and other factors (ACR, 1993). Larger magnifications are likely to result in increased dose and excessive geometric blur, with associated motion unsharpness along with decreased field size in the breast.

The nominal size of the focal spot used for magnification should be 0.10 mm or less (Eklund and Cardenosa, 1992).

Since grids cannot generally be used in conjunction with magnification, it might be expected that magnification images lack sufficient contrast. Two factors prevent this from being the case. First, the air gap, the separation introduced between the exit surface of the breast and the image receptor, reduces the amount of scattered radiation detected by the image receptor. Rather than absorbing scattered radiation as is the case with a grid, the air gap provides an opportunity for scattered radiation to project off the film. The greater the air gap, the less scattered radiation will be recorded. Second, coning down the x-ray field size to as small an area as possible will limit the production of scattered radiation (Hall, 1989; Sickles, 1989). Scattered-radiation production is also limited by firm compression as it is in contact mammography. Compression is also critical for the other reasons mentioned for contact mammog-raphy, particularly in helping to prevent even the slightest motion during the relatively long exposure times typical of magnification mammography (Sickles, 1987a).

Spot compression coupled with magnification has been demonstrated to be particularly effective (Faulk and Sickles, 1992). A round or narrow compression device, no wider than 9 cm is essential for firm compression of the area to be magnified. As mentioned previously (Section 2.1.8), devices are also available which provide a small raised area on the patient support corresponding to the size and location of the spot-compression device and thus provide reciprocal compression from below as well as above.

Mammographic units used for magnification should be equipped with collimation systems that allow either exposure of the entire image-receptor or coned-down views; the light field should indicate clearly the location and size of the x-ray field (ACR, 1993). As mentioned above, a magnification stand is also necessary. This device is attached to the usual image-receptor support and allows for positioning the breast considerably above the plane of the image receptor. When imaging is performed in this elevated position, the dose to the patient is increased. However, assuming that a grid is not used (which will be true in virtually all cases), the dose will not be significantly increased compared to standard non-magnification grid imaging. At the same time, the breast image is magnified improving image resolution. The magnification stand should be easy to set up; otherwise magnification can become a burdensome chore.

The magnification stand should remain solidly in place and not slant downward with continued use, a downward slant means that the stand is no longer parallel to the image receptor and the compression device. The chest-wall edge of the magnification stand should align perfectly with a line between the focal spot and the posterior edge of the film. An abdominal shield should also be provided to prevent the patient's abdomen from projecting between the breast and the image receptor. Such an abdominal shield should be sturdy and rigid so that it can neither break easily nor be pushed into the x-ray field.

The unit should be designed so that the technologist can lower the C-arm enough to perform magnified views on short patients or those who must be imaged in the seated position. The space between the breast support and the tube head must be large enough to allow good positioning. If not, the patient's shoulder will hit the tube head and the technologist will not be able to pull all of the patient's breast onto the magnification stand. If the entire breast is not on the stand, the radiologist will not be able to visualize lesions at the chest wall. The unit should contain diaphragms matching the width of all the compression devices including the small compression device for the coned-down view. Table 3.14 presents the desirable characteristics of magnification mammography.

Table 3.14—Desirable characteristics for magnification mammography.

• Magnification

- 1.5 to 2 times magnification available

- Magnification stand provided and easy to set up

- Magnification stand solidly attached

- No downward slant with use

- 0.1 mm nominal focal spot

• Miscellaneous criteria

- Chest-wall edge aligned with posterior edge of film

- Sturdy, rigid abdominal shield available

- Sufficient space between breast support and tube head to allow good positioning

- Round or narrow compression device

- No wider than 9 cm for spot compression

- Collimation matching width of all compression devices

0 0

Post a comment