Dedicated mammography imaging systems were first developed in 1970s, and made use of x-ray tubes with molybdenum targets and filters made of molybdenum (Z = 42) or rhodium (Z = 45). Mammography also uses low x-ray tube voltages, typically 25 kV or so, x-ray tube currents of 100 mA, and exposure times of 1 second or more depending on the thickness of the compressed breast. An average sized breast with 50% glandularity would require x-ray techniques of 25 kV/100 mAs, and result in an average glandular dose of 2 mGy; the ACR accreditation program requires that the average glandular dose to this sized breast to be less than 3 mGy.
Screen film mammography is currently being replaced by digital systems, which offer operators the ability to modify x-ray radiographic techniques (kV, mAs, etc.) over a wider range of values than analog screen film systems. The overall goal of digital mammography is to offer users high image quality on a consistent basis, whilst keeping patients doses relatively low.
One important feature of any digital imaging system is the ability to use different amounts of radiation whilst still generating an image that has a satisfactory gray scale appearance. All the examples depicted below were generated using a digital mammography system that records an exposure index (EI) value that is a measure of an average intensity incident on the digital detector in a clinically relevant region as determined by the system manufacturer. Low values of EI correspond to the use of low radiation intensities (low patient doses), and vice versa. Another characteristic of digital mammography systems is the ability to modify the x-ray beam quality (i.e., penetrating power or average photon energy) by modification of the x-ray tube voltage (kV) and the filter material (Mo or Rh). In general, increasing the x-ray kV, and/or using a filter with a higher atomic number (Z = 42 for Mo; Z = 45 for Rh) increases the x-ray beam penetrating power (average energy) that will reduce the breast dose. The reason for this reduction in patient radiation dose with increasing photon energy is that less radiation is needed incident on the breast when the receptor dose is kept constant.
In the cases presented below, we use of two types of phantom to generate images that may be used to assess image quality. In addition, detailed explanations are provided as to how the exposure protocol impacts on the corresponding average glandular dose. These phantoms are:
A. Anthropomorphic phantom that has been manufactured to simulate images that simulates a realistic breast. This phantom is frequently used for research applications, and permits radiographic techniques to be varied without actually irradiating any patient(s). The composition of the anthropomorphic breast phantom, however, does not reflect a true breast, but merely ensures that the transmitted x-ray intensity pattern reflects that of a typical mammogram. Accordingly, all dose estimates for anthropomorphic phantoms should always be interpreted as relative values (i.e., not absolute values). In other words, the anthropomorphic breast phantom may be used to illustrate how doses change when factors such as kV and/or filter are altered, but will not accurately predict the absolute dose level for any given patient.
B. ACR accreditation image quality phantom that has a composition of an average sized (compressed) breast with 50% glandularity. The composition of this ACR image quality phantom is such that it will accurately predict the average glandular dose to an average sized breast, and is normally used to ensure that the dose delivered by the mammography system meets the required limit (i.e., < 3 mGy). This ACR phantom also contains inserts consisting of simulated fibers, microcalcification clusters, and masses; this image quality phantom, which is used on a weekly basis in most mammography facilities, permits an objective assessment of mammography image quality based on the visibility of these lesions that reflect the features of interest in clinical mammography.
The digital mammography system uses a very short test exposure to estimate the breast composition based on the transmitted intensity, and also makes use of explicit knowledge of the compressed breast thickness from a sensor incorporated into the breast compression device. The system can be operated in one of the following four modes:
Auto Filter. This mode is used to identify the optimal filter material (Mo or Rh) together with the optimal x-ray tube voltage.
Auto kV. If the Filter is manually set, the system identifies the optimal kV.
Auto Timing. If both the filter and kV are manually set, the system will terminate the exposure when there is a preset level of radiation incident on the detector (i.e., the mAs).
Manual. In this mode, all the parameters (i.e., filter, kV, mAs) are selected by the operator, and any desired amount of radiation may be used. When the digital system is operated in manual mode, it is possible to illustrate the differences in filter material, x-ray tube voltage, as well as the total amount of radiation used (mAs).
The amount of radiation incident on the detector is quantified by an exposure index (EI) parameter, and this value is directly proportional to the total mAs value used in any given exposure. Since the tube current is normally fixed at 100 mA, the mAs is essentially an indicator of the exposure time (i.e., 100 mAs corresponds to a 1 second exposure). For the imaging system used to acquire images of the two phantoms described above, the target design is an average glandular dose of ~ 1.5 mGy that corresponds to an Exposure Index value of ~ 450. In each Auto mode (i.e., Auto filter; Auto kV; Auto Timing), the exposure time is determined by how long it takes to reach this EI value of ~450.