The mammography system: why it is different

Name the blocks of a mammography system, explain why each one is built differently from a general X-ray unit, and recognise which detector technology you are standing in front of.

Same physics, different job

A mammography unit is an X-ray system, and everything you learned in X-ray Service Level 1 about the generator, the tube, the collimator and radiation safety still applies. What changes is the job. The breast is made of soft tissues whose X-ray attenuation is very close together: fat, glandular tissue and most tumours differ only slightly. On top of that the radiologist is looking for microcalcifications, specks of calcium a fraction of a millimetre across. A general radiography unit set up for chests and bones would show a grey, flat picture with none of that detail.

So every block of the system is tuned for two things at once: high contrast between similar soft tissues and very fine spatial detail, while keeping the dose to the glandular tissue as low as possible. Because many women are screened who have no disease at all, dose matters even more than in diagnostic radiography. When you understand those three goals, contrast, detail and dose, almost every design choice in the machine makes sense, and so do most of the faults that radiologists complain about.

Low kV, special targets and filters

Contrast between soft tissues is best at low photon energies, where the photoelectric effect dominates. That is why mammography runs at roughly 25 to 35 kV instead of the 50 to 150 kV used elsewhere. At these energies a few kilovolts make a big difference to both contrast and dose, so kV accuracy and the choice of target and filter matter far more than on a general room.

Classic units use a molybdenum anode with a molybdenum filter. Molybdenum emits characteristic X-rays in a narrow energy band that suits an average breast, and a molybdenum filter of the same element removes most of the low-energy photons that would only add dose, and much of the high-energy tail that would wash out contrast. A rhodium filter, or a rhodium anode, shifts the spectrum slightly higher for denser or thicker breasts. Modern digital systems often use a tungsten anode with rhodium or silver filters: a digital detector can restore contrast in processing, so the system can use a slightly harder beam and save dose. You will see combinations written as Mo/Mo, Mo/Rh, Rh/Rh, W/Rh or W/Ag on the console and in the exposure log.

The tube window is beryllium rather than glass, because glass would absorb the very soft X-rays the system needs. The filter assembly is usually a small motorised wheel or slide in front of the window, and the AEC picks the target and filter combination for each breast. A filter that is not where the system thinks it is gives the wrong spectrum, and on many units it is a real error message, not just a quality issue.

Small focal spots, short distances, and a grid built for soft X-rays

Fine detail needs a small focal spot. Mammography tubes normally have two: a large focus, often around 0.3 mm nominal, for contact views, and a small focus, around 0.1 mm, for magnification views where the breast is lifted away from the detector on a stand. The GE Senographe Crystal Nova documentation, as one example, lists 0.3 and 0.1 mm foci and a source-to-image distance of about 66 cm. The tube is usually tilted so that the central ray runs along the chest wall: the side of the image next to the patient's body must not be cut off, because tissue there is lost from the picture.

Scatter reduces contrast, so most units use an anti-scatter grid inside the breast support, the bucky. A mammography grid has a low ratio and thin, low-absorbing interspace material such as carbon fibre, because the beam is so soft. Most grids move during the exposure so that the grid lines are blurred out; a grid that stops, sticks or starts late leaves fine lines or a pattern on the image, or blocks the exposure altogether if the system checks the grid position. Magnification views are taken with the grid out, because the air gap does the scatter rejection.

Automatic exposure control in mammography is not optional. Breasts differ hugely in size and density, and the radiographer cannot guess the right exposure. Older units used a separate AEC detector behind the image receptor; digital units usually read a short pre-exposure from the main detector and then choose kV, target, filter and mAs. Lesson 3 covers how it behaves and fails.

Gantry, C-arm, compression and the workstation

The gantry is a column with a motorised vertical drive carrying a C-arm. The tube head sits at the top of the C-arm and the breast support with detector and grid at the bottom. The C-arm rotates so that the radiographer can take the standard views, cranio-caudal straight down and medio-lateral oblique at an angle, on both sides. Between tube and detector runs the compression paddle, driven by a motor and stopped by a force sensor. Compression is covered in lesson 2; for now remember it is part of the imaging chain, not just a holder, because it spreads tissue, reduces thickness and with it scatter and dose, and keeps the breast still.

A face shield protects the patient's head from the primary beam and keeps her face out of the image. Footswitches and keypads on both sides of the C-arm move the arm and the paddle. Emergency stop buttons are fitted on the gantry and at the console, and on many units pressing one also releases compression.

The acquisition workstation, behind a shielded screen, lets the radiographer select the patient, set the exposure mode and check every image before sending it to PACS. It is a PC with an operating system, disks, a network link and often a separate detector or image-processing computer. A large share of real mammography faults, as you will see in lesson 6, live here rather than in the X-ray hardware. The reading station the radiologist uses is a separate system with high-brightness, high-resolution monitors that have their own quality control.

Detectors: film, CR, DR and tomosynthesis

You will still meet every generation of image receptor in the field, sometimes in the same hospital, so learn to recognise them quickly.

Screen-film mammography uses a single-emulsion film in a cassette with one intensifying screen, processed in a film processor with chemistry made for mammography film. Image quality depends as much on the processor as on the X-ray unit: temperature, chemistry type, replenishment and the darkroom. CR uses cassettes holding a storage phosphor plate that is read in a separate reader; mammography CR plates and readers are higher resolution than general ones, and the reader settings must match the cassette size and type. DR, full-field digital mammography, has a flat-panel detector built into the bucky. Two main technologies exist. Direct conversion panels use amorphous selenium, which turns X-rays straight into charge; indirect panels use a caesium iodide scintillator on an amorphous silicon photodiode array, which turns X-rays into light first. The Crystal Nova, for example, uses CsI on amorphous silicon with a 100 µm pixel pitch. Some systems, such as photon-counting scanning designs, work differently again.

Digital breast tomosynthesis takes a series of low-dose exposures while the tube moves through a small arc, and reconstructs thin slices. For the engineer it adds a moving tube head with its own drive and position sensing, more data per study, and heavier demands on the detector readout, the workstation disks and the network. At Level 1 you only need to know it exists, that it is more sensitive to mechanical play and timing, and that its faults are often data and storage faults.

• Film: single-emulsion film, one screen, mammography chemistry, processor QC is half the job.
• CR: high-resolution plates and reader, settings must match the cassette.
• DR direct: amorphous selenium, X-rays to charge.
• DR indirect: CsI scintillator on amorphous silicon, X-rays to light to charge.
• Tomosynthesis: moving tube, many low-dose frames, big data load.

Your first look at an unfamiliar unit

When you walk up to a mammography unit you have not worked on, spend five minutes identifying what you have before touching anything. Read the label on the gantry and on the generator; find the main switch, the emergency stops and the way to release compression by hand. Look at the console: does it show target and filter, thickness and force? Find out whether the receptor is film, CR or DR, and if DR, which detector and how it is cooled and powered. Find the service and user documentation for that model and the department's QC records.

Then ask the radiographer three questions: what exactly is wrong, on which views and modes, and since when. Mammography faults often show on one mode only, for example magnification, small focus, tomosynthesis or one target and filter, and that is the strongest clue you will get.

On-site checklist

• Identify the receptor type: film, CR or DR (and which detector technology).
• Find main switch, emergency stops and the manual compression release before working.
• Note the target and filter combinations the unit offers and what the console shows.
• Locate the model's user and service documentation and the department's QC records.
• Ask which views, modes and foci show the problem, and since when.

Check yourself