Phantom Thickness (cm)
Fig. 3.9. Breast thickness tracking with typical AEC devices on four different (A through D) commercially available mammography units (Barnes, 1999).
the microprocessor controlled program of the AEC. This allows the variation of the AEC sensor sensitivity with operating potential to be taken into account in determining when the exposure should be terminated [e.g., based on a table with different threshold values for each operating potential (Kimme-Smith, 1992)]. Variations in compressed breast thickness also result in beam hardening for which corrections must be made. The AEC can, for example, be equipped to evaluate the energy of the x-ray beam exiting the breast. Thicker and denser breasts are more attenuating and will result in a higher average exit beam energy. If the exit beam energy is known, then the unit can correct for the variation of the AEC detector sensitivity with energy by making an appropriate adjustment in the threshold.
Thick or dense breasts or both also create problems because long exposure times result in reduced image-receptor response due to film reciprocity law failure. Modern AECs provide mechanisms to compensate for the effects of reciprocity law failure, typically in the unit's software.
Independent of the compensation method employed (including systems using automatic operating potential selection), the AEC should insure the production of uniform optical density images independent of breast thickness and operating potential. For a set of images made with tissue thicknesses of 2 to 6 cm and for the range of operating potential settings appropriate for those thicknesses, the optical densities should not vary by >0.15 optical density from the mean optical density of the set (ACR, 1999). This performance should be evaluated with a mean optical density >1.2. The type of film and processing should be specified, since the film characteristics will affect the degree of optical density change due to a fixed difference in exposure. Furthermore, the AEC should meet the same standard for a phantom thickness simulating an average breast thickness when imaged over the entire range of operating potential settings used clinically.
Equally important, AEC must be reproducible. It is not difficult for recently manufactured systems to meet the requirements of the federal performance standard which requires that the coefficient of variation of a set of exposures not exceed five percent (ACR, 1999; AHCPR, 1994). Exposures reproducible to this level should be possible in the AEC mode between 5 and 300 mAs.
At least three sensor positions (or multiple sensors) on the AEC should be provided (Feig, 1987; Logan, 1983). This range of positions allows the technologist to adjust for variations in individual patient anatomy, choosing the best position for placing the detector. This makes it possible to optimize the exposure to critical regions where pathology is most likely to be found. The field of the AEC detector should be large enough that a representative amount of breast tissue is sampled, but not so large that it would not be completely covered by a small breast (ACR, 1993; Feig, 1987). If the detector extends beyond the breast, part of the detector will be exposed to unattenuated x rays. Under such conditions, the detector will reach its threshold too soon and will terminate the exposure too early, resulting in an underexposed image. Some newer AECs can average the density of the tissue throughout the entire breast, which makes positioning the AEC less critical (Kimme-Smith, 1992). The potential position and size of the AEC detector should be clearly indicated at the top surface of the breast (ACR, 1993), usually on the compression device. Ideally, the position of the AEC detector should be continuously variable along a line oriented in the chest wall to nipple direction.
The AEC system should also be provided with an optical density adjustment with at least nine clearly indicated density adjustment steps (ACR, 1993). There should be at least four steps above and below the normal density setting. Each step should alter the milliampere seconds by approximately 10 to 15 percent from the adjacent step. This is a somewhat more restrictive specification than the 15 to 20 percent recommended a few years ago (AAPM, 1990). The change was necessitated by the fact that step-to-step increments as small as 12.5 percent may sometimes be too large for fine density adjustment due to the high contrast of some mammo-graphic films. There should also be adjustments provided so that the AEC system can be set appropriately for different screen-film combinations (AAPM, 1990).
The unit should offer the technologist the option of either setting the operating potential before positioning or permitting the x-ray unit to choose the operating potential. In the latter case, operating potential control is accomplished by beginning the exposure at a predefined kilovolt peak and then using the AEC to evaluate the exposure rate at the sensor location. If the breast is highly attenuating, the exposure rate will be low and the unit will increase the operating potential to achieve an exposure rate that will result in appropriate optical densities in a reasonable exposure time (Barnes, 1994). If the unit automatically selects the kilovolt peak and adjusts it during the exposure, there should be an accurate postexposure display of the actual kilovolt peak employed. If the technologist chooses to set the operating potential, the unit should not be able to override the technologist's selection.
When the unit is switched to AEC, the technique used most frequently should automatically be set as the default technique and these factors should be indicated on the control panel. If the preset technique is inadequate, it should be impossible to make an exposure (unless the unit is equipped with the auto operating potential feature discussed above and the unit is operating in that mode). Ideally, the AEC device should be capable of determining whether the back-up time is likely to be reached and, if so, should terminate the exposure within 50 ms, 5 mAs, or 13 pC kg1 and indicate the termination to the technologist (ACR, 1993). Alternatively the system should, under such conditions, increase the operating potential so that the exposure can be made in a reasonable time. Long exposures such as those that would nearly reach the back-up time should be avoided because they are subject to motion artifacts and may need to be retaken. Exposures that reach the back-up time will be underexposed and also need to be retaken. In both cases, the patient dose is increased unnecessarily.
An indicator displaying the postexposure milliampere seconds should be provided and the displayed milliampere seconds should be held, or be retrievable, until the next exposure (AAPM, 1990; ACR, 1993). Dose estimation for individual patients is greatly facilitated by this type of display, which also assists in technique selection for manual exposures and retakes. The unit should also incorporate a back-up timer to limit the exposure in case of a system failure. If there is such a failure, the unit should indicate that the back-up time (or milliampere seconds) was reached. The backup time should provide user selectable settings, but it should not be possible to set it below 250 mAs for contact mammography (ACR, 1993). Due to focal-spot loading considerations, it may be appropriate to have a 50 mAs lower limit for microfocal-spot tubes used in magnification mammography. The maximum limit allowed by the federal performance standard is 2,000 mAs for general radiography. A maximum limit of 600 mAs is more appropriate for mam-mography (ACR, 1993). In some designs the last manual exposure time is used as the backup and this can cause a problem, sometimes resulting in an underexposed film for which a retake is necessary. Table 3.11 presents desirable characteristics of exposure control devices (automatic and manual).
126.96.36.199 Manual Exposure Controls. Manual exposure controls are also essential, particularly for imaging patients with implants, for special views, for specimen radiography, and for certain QC tests. Manual exposure time or milliampere-seconds selections should range from 0.02 to 6 s (2 to 600 mAs at 100 mA) in 15 to 20 percent increments (AAPM, 1990; Yaffe, 1991). All time or milliampere-seconds selections should result in reproducible exposures with a coefficient of variation of <5 percent (CDRH, 2002b) from 5 to 300 mAs (ACR, 1993). Indicators displaying the preset milliampere seconds should also be provided in addition to the postexposure milliampere-seconds display. The radiation output of the unit using manual exposure control factors should be within five percent of that in the AEC mode for the same operating potential and postexposure milliampere seconds (ACR, 1993).
TABLE 3.11—Desirable characteristics of exposure control devices.
3.1.8 Compression Device
Firm compression is essential in mammography for a variety of important reasons: reduces geometric unsharpness, reduces scattered radiation, diminishes motion unsharpness, reduces x-ray dose, produces more uniform film density, accentuates the differences in density between normal and malignant tissue, and separates overlapping tissue elements. For all these reasons, a properly designed compression device must be provided on a mammographic x-ray unit (AAPM, 1990; ACR, 1993).
The impact of compression on scattered radiation may seem exaggerated. After all, the breast is not really "compressed," but simply spread out over a larger area. The same tissue is exposed and thus, the same volume of tissue is producing scattered radiation. However, the production of scattered radiation increases much more rapidly with increasing thickness than it does with increasing field size (Figure 3.10) (Barnes, 1994). Therefore, the breast thickness reduction achieved by firm compression results in a significant decrease in scattered radiation production, in spite of the increase in breast area. In addition, by reducing breast thickness, compression also reduces beam hardening which also improves contrast. In the absence of scattered radiation and with a Mo/Mo unit operated at 28 kVp, the contrast of microcalcifications increases seven percent per centimeter of decrease in compressed breast thickness (Wagner, 1991).
Proper compression device design (see below) is essential if adequate compression is to be achieved without the patient experiencing undue discomfort. However, as noted elsewhere in this Report (Section 2.5.4) (Eklund, 1991), there are a wide variety of factors that affect the patient's experience of compression, not least of which is the skill and sensitivity of the technologist. It is impossible to overestimate the technologist's role in achieving adequate compression and all the benefits that result.
The compression device should be an integral part of the x-ray unit, mounted rigidly, so that it may be positioned in a reproducible fashion (ACR, 1993; AHCPR, 1994). This will facilitate proper positioning and firm compression of all the breast tissue. A stiff compression device, which is perfectly flat and parallel to the image-receptor surface, should be utilized (AAPM, 1990; ACR, 1993; Feig, 1987; NCRP, 1986; Yaffe, 1991). It is important that the compression device remain as nearly flat and parallel to the image receptor as possible during compression. If the compression device does not remain parallel to the image receptor during compression, but rather slopes posteriorly, tissue at the base of the breast will be less compressed and relatively underpenetrated and many of the benefits of compression will be lost, particularly in the posterior regions of the breast. Unfortunately, early compression devices were designed to slope posteriorly. The advantages of a flat design were not recognized until the late 1970s (Logan and Norlund, 1979).
Was this article helpful?