Imaging Objectives

Rhodium Anode

Blurring and Visibility of Detail

Visibility of Pathology

KV Values for Mammography

Focal Spot Blurring

Image Quality Characteristics

Scattered Radiation and Contrast

Receptor Blurring

Not a Perfect Image

Image Exposure Histogram

Composite Blurring

Mammography Technology

Receptor & Display Systems

Magnification Mammography

Imaging Technique Factors

Film Contrast Transfer

Mean Glandular Dose

Contrast Sensitivity

Film Contrast Factors

Physical Contrast Compared

Film Design for Mammography

Factors Affecting Contrast Sensitivity

Controlling Receptor (Film) Exposure

X-Ray Penetration and Contrast

Film Processing

Optimum X-Ray Spectrum

Variations in Receptor Sensitivity

Effect of Breast Size

Film Viewing Conditions

Controlling the X-Ray Spectrum

Digital Mammography Contrast

The Molybdenum Spectrum

Moly and Rhodium Filters

Let's begin with a simple observation that will be a foundation for a much more in-depth study of the mammography imaging process. We can all agree on the objectives of the procedure.  The challenge is in achieving the desired results.  It begins with the fact that many pathologic conditions, especially cancer, produce very small physical changes that are difficult to visualize with  x-ray imaging.

This is what makes mammography the examination  that requires the highest image quality of all of the x-ray procedures.
Maximum visibility, especially of the signs of pathology, is achieved by using state-of-the -art equipment and imaging protocols (technique factors, image processing, etc.) that optimize the procedure and balance the quality requirements with the radiation dose to the patient.

This requires an in-depth working knowledge of the applied physics and technology by the clinical staff.

Saying that visibility of pathologic conditions depends on image quality sounds like a "no Deterrent."  It is a simple point but the complexity comes from the combined factors that the various signs of pathology have very different physical characteristics and that image quality is not one, but a combination of several specific image characteristics.

Most signs of breast pathology are either in the form of soft-tissue masses that are not very different from the surrounding tissue or in the form of very small (micro) calcifications.

As we will see, the visualization of these very challenging conditions requires imaging procedures with special characteristics.

The quality of a mammogram (just like any other medical image) is not one single characteristic.  It is a composite of five very specific characteristics as we see here.

Visibility of anatomy and pathology, the primary objective of mammography, can be affected by each of these individual characteristics, but in very different ways.

The design of equipment, imaging protocols and techniques, and quality assurance procedures generally address each of these characteristics on an individual basis.

When there is a suspected image quality problem a first step is to identify which one of the five characteristics is the source and then do an evaluation that will lead to an appropriate corrective action.

We will consider each of these characteristics in detail as we work through the imaging process.

The goal in clinical imaging is to adjust the imaging factors and protocol to extend the range of structures and objects in a breast that will be visible.  However, we must always be aware that there is no perfect image that will show every possible pathologic feature.

The limitation of visibility with mammography is routinely evaluated by imaging the accreditation phantom illustrated here.

The phantom is a cube of material designed to simulate "average" breast tissue compressed to a 4.5 cm thickness.

There are three (3) sets of objects within the phantom to simulate structures and objects that might be within a breast.  These are the "masses, fibers, and specks."  We will take a closer look at each of these a little later in the context of the characteristics that they are used to evaluate.  What is common to all three types of objects is that they cover a range of sizes.  The large objects (masses, fibers, or specks) will always be visible, even under inferior imaging conditions.  A mammography system and a specific imaging protocol is evaluated by imaging the phantom and then counting the number of objects of each type that are visible.

In this illustration (on the left) we see a diagram of the phantom interior showing the various objects that are included.  On the right we see an image of the phantom and note that all of the objects are not visible.  NOTE: this is a very small image and not typical of an actual mammogram but is used here to demonstrate an important point.  Even in mammography, there is generally not a perfect image providing visibility of all possible structures and objects within a breast.

We will use the phantom again later to evaluate specific image characteristics.   
 

Equipment for mammography has evolved over at least the last 40 years to the current state of the art.  While there are some differences from one manufacturer to another, there are also many characteristics and features that are common to all.  That is what we will introduce here and then cover in more detail later.

X-ray Tube Anode: Whereas most x-ray tubes use tungsten as the anode material, mammography equipment uses molybdenum anodes or in some designs, a dual material anode with an additional rhodium track.  These materials are used because they produce a characteristic radiation spectrum that is close to optimum for breast imaging as described later.

Filter: Whereas most x-ray machines use aluminum or "aluminum equivalent" to filter the x-ray beam to reduce unnecessary exposure to the patient, mammography uses filters that work on a different principle and are used to enhance contrast sensitivity.  Molybdenum (same as in the anode) is the standard filter material. Some systems allow the operator (or automatic control function) to select either the molybdenum or a rhodium filter to optimize the spectrum for specific breast conditions.

Focal Spots:  The typical x-ray tube for mammography has two selectable focal spots.  The spots are generally smaller than for other x-ray procedures because of the requirements for minimal blurring and good visibility of detail to see the small calcifications.  The smaller of the two spots is generally used for the magnification technique.

Compression:  Good compression of the breast is one of the essentials of effective mammography (and a common source of patient discomfort and concern).  Potential benefits derived from compression include:

1. A more uniform breast thickness resulting in a better fit of the exposure into the film latitude or dynamic range.

2. Reduced blurring from patient motion.

3. Reduced scattered radiation and improved contrast sensitivity.

4. Reduced radiation dose.

5. Better visualization of tissues near the chest wall.

Grid:  A grid is used in mammography (as in other x-ray procedures) to absorb scattered radiation and improve contrast sensitivity.  Compared to grids for general x-ray imaging, grids for mammography have a lower ratio and the material between the strips is selected for low x-ray absorption.

The grid is contained in a Bucky device that moves it during the x-ray exposure to blur and reduce the visibility of the grid lines.

Receptor:  Both film/screen and digital receptors are used for mammography.  Each has special characteristics to enhance image quality and will be considered later.

Optimizing a mammography procedure for maximum visualization of anatomy and signs of pathology without unnecessary radiation to the patient is achieved by the selection of the best combination of technique factors that make up the imaging protocol.

Each of these factors generally has an effect on one of the specific image quality characteristics and in many cases, an effect on patient radiation dose.  At this point we identify the principal technique factors and their general functions.  The detailed applications will come later.

X-ray Beam Spectrum:  The photon energy spectrum of the x-ray beam is one of the most critical factors in optimizing a procedure with respect to contrast sensitivity and radiation dose.  The spectrum depends on the combination of three factors:

1. The anode material (molybdenum or rhodium)

2. The selected filter   (molybdenum or rhodium)

3. The KV (ranging from 24kV to 32kV)

These factors are either set manually by the technologist/radiographer after evaluating breast thickness and density or by the automatic exposure control (AEC) function if it is available.  The AEC typically makes a brief exposure to measure penetration through the breast and from that calculates appropriate technique factors for the imaging.

When using the AEC to automatically set the spectrum the operator also has a role in optimizing the technique factors.  Most equipment with this feature allows the operator to select a "priority" for the procedure in the balance between contrast sensitivity and radiation dose.  When the "CONTRAST" mode is selected the spectrum controlling factors (Anode, Filter, and KV) are automatically set to produce a spectrum that enhances contrast sensitivity.  When the "DOSE" mode is selected the factors are automatically set to produce a more penetrating spectrum that reduces the dose but at the cost of reduced contrast sensitivity.  The relationship between contrast sensitivity and dose is described later with an emphasis on the need to optimize the imaging factors for specific breast conditions.

Receptor Exposure Control:  Image quality depends on the appropriate exposure being delivered to the receptor.  This applies to both film/screen and digital receptors but for different reasons. With film/screen the objective is to expose the receptor to a level that will produce the film density that gives maximum contrast.  With digital receptors the contrast generally does not depend on the level of receptor exposure but the exposure must still be carefully controlled to optimize the procedure with respect to image noise and radiation dose to the breast.

The AEC system is useful in obtaining optimum exposure but does require involvement of the operator.  First, the AEC sensors must be selected and positioned in relation to the anatomy and breast conditions and the "Density" or exposure control set to optimize for variable conditions.

The AEC system must be periodically evaluated by a medical physicist and set-up by the engineering staff to insure proper performance.

A major requirement for effective mammography is high contrast sensitivity.  As illustrated here, contrast sensitivity is the characteristic of an imaging process that determines the visibility of objects in the body that have low physical contrast.  That is the challenge we have in mammography. The anatomical structures and pathologic signs are all soft tissues with physical densities very similar to the adipose background of the breast.  The visibility of small calcifications is limited by blurring and will be discussed later, but they also require a high contrast sensitivity.

Contrast sensitivity is a fundamental characteristic of an imaging system and the imaging procedures with each system.

The different imaging modalities (radiography, CT, MRI, etc.) have different contrast sensitivities because of their design and physical principles.  Of the x-ray imaging modalities, CT has the highest contrast sensitivity for visualizing soft tissues.  For the radiographic imaging methods, mammography is designed to have a much higher contrast sensitivity than the other radiographic procedures.

The relative contrast sensitivity is one of the characteristics that is tested using the accreditation phantom shown here.  That is done by counting the number of simulated masses that are visible.  The phantom contains a series of five (5) simulated masses.  We see them as decreasing in size (diameter) from the largest (#12) to the smallest (#16) in the lower right corner.  However, it is not the diameter that is important; it is the thickness of the masses which is also decreasing as shown here.

The thickness of a mass determines its physical contrast and the amount of x-ray attenuation it produces.  Since we have a series of objects (masses) with varying physical contrasts, it is a useful test device for evaluating contrast sensitivity.

It is expected to be able to see four (4) masses when the phantom is imaged with conventional mammography equipment operating in a standard protocol.  The fifth mass is generally not visible.  This does not indicate that the equipment or imaging procedure is defective.  It does demonstrate to us that even with high-quality mammography, every thing within a breast might not be visualize.

The one disk not in the numbered series (red arrow) is actually a small disk attached to the outside of the phantom and used to measure overall contrast in the image using a densitometer. A numerical value for the contrast is the difference between the film density values measured in the disk area and the background area near the disk.

As we will soon see, there are a number of technical factors within mammography that affect contrast sensitivity.  These need to be considered when setting up the procedure, processing,  and viewing the image.

The level of contrast sensitivity that is needed in a specific imaging procedure depends on the amount of physical contrast that is present in the body section being imaged.  That varies considerably among the different anatomical locations.

It is interesting to consider the two extremes illustrated here.  The chest is a region with very high physical contrast because of the large difference in density between the lungs partially  filled with air and the bones. The lungs form a low density background on which most of the other anatomical structures and signs of pathology can be imaged.

Chest radiography requires low contrast sensitivity because of the high physical contrast that is present.  The first step to achieve low contrast sensitivity is by using  high KV values (like 120kV)  that produce a very penetrating x-ray beam.

The breast is the complete opposite to the chest with respect to contrast.  It consist ofs soft tissues with relatively small differences in density (or atomic number).

The adipose tissue does form a "low density" background on which the glandular tissue and signs of pathology can be imaged.

However, the differences in density and the physical contrast is very small and a procedure with high contrast sensitivity is required for visualization.

We now move on to how high contrast sensitivity is achieved in mammography.

The breast imaging process consists of a sequence of actions and events transferring the physical contrast in the breast to visible contrast in the displayed image.  There are factors associated with each of these that have an effect on contrast sensitivity.  We will introduce them here and then go into more detail as we work through the imaging process.

Anatomical Environment: This is where it all begins.  As we have just seen the challenge in mammography is to visualize the very low physical contrast in the breast.  The natural shape of the breast with varying thickness from the nipple to the chest wall is a general deterrent to achieving good contrast and visibility.  Good compression transforms the breast into a more uniform thickness and a somewhat thinner environment for better imaging. 
Dense breasts are more difficult to image because structures of interest are contained in the more dense tissues rather than in an adipose background.

The X-ray Beam Spectrum:  One of the most unique features of mammography compared to other radiographic procedures is the x-ray beam spectrum that is used.  A spectrum with relatively low photon energies is required to produce the high contrast sensitivity and is adjusted to specific breast conditions to optimize contrast with respect to radiation dose to the breast.

Scattered Radiation:  Even though the breast is small, compared to other body sections, it is still a source of scattered radiation that reduces contrast.  Compression has some effect on reducing the scatter.  The scatter reaching the receptor is reduced with a grid designed specifically for mammography.  The grid is generally not used with the magnification technique and the air gap decreases the intensity of the scatter reaching the receptor.

Film:  One of the functions of film is to transfer the contrast contained in the x-ray beam emerging from the breast into visible contrast in the displayed image.  There are three (3) major factors that affect this transfer.  They are:

• Design Characteristics of the Film

• Level and Range of Exposure Delivered to the Film.

• The Quality of the Chemical Processing

Each will be considered in detail later.

Film and Image Viewing Conditions: The display and viewing of the mammogram is the last step in the visualization of the breast.  Because of the requirements for high contrast sensitivity and the good visualization of detail (the small calcifications) optimum viewing conditions are required.

An x-ray image is a shadow of the internal structures and objects within the body.  A shadow is produced when an object absorbs or attenuates some of the x-ray beam as it passes through.  The contrast in the shadow image is reduced with increased penetration through a specific object as illustrated here.

Using the light analogy, a clear piece of glass produces a low contrast shadow because most of the light passes through or penetrates it.

In x-ray imaging the contrast produced by a specific object is increased by decreasing the penetration.  For a specific object, the degree of penetration and resulting contrast depends on the photon-energy spectrum of the x-ray beam.

The penetration through soft tissues and fluids found in the body (that have relatively low atomic numbers, Z values) generally increases with increasing photon energy.  Therefore, contrast is increased by using an x-ray spectrum consisting of low-energy photons.

However, there is another factor to consider...that is the penetration through the total body section, or breast in the case of  mammography.
  Generally we want maximum penetration through the breast for two reasons. With good penetration less radiation is required into the breast to produce the desired receptor exposure.  This reduces the amount of radiation and heat that must be produced by the x-ray tube for each exposure.  The benefit is that the exposure times can be kept relatively short to reduce motion blurring (a problem especially with large breasts) and less heat is produced in the focal spot area.  This is a potential limiting factor when using the small focal spot for the magnification technique.

The major concern associated with penetration is the radiation dose to the breast.  The radiation energy, or dose, that must be deposited in the breast is increased as the penetration is decreased.

Therefore, in mammography we have two conflicting requirements. W e need relatively low penetration to enhance contrast but high penetration is needed to reduce the dose to the breast.

The solution is to use an x-ray beam that has a spectrum that produces an optimum balance between the requirements for high contrast sensitivity and low radiation dose.    

Let's do a quick review of the interactions of x-ray photons with tissue.  At the relatively low photon energies used for mammography many of the interactions are by the photoelectric process.  The significance is that the rate of photoelectric interactions depends on the energy of the individual photons, and generally decreases with increasing energy.  This means that both dose and contrast decrease with increasing photon energy as shown here.  If they should both decrease at exactly the same rate we would be out of luck!  Fortunately because of some variation in composition of the breast and some effect of thickness, the dose and contrast do not decrease at the same rate and that is what makes it possible to adjust the x-ray beam spectrum for optimized imaging.

If we could measure the contrast and dose and determine the contrast to dose relationship for a specific breast size and density it would look somewhat like the curve shown here.  There is a "peak" at some specific photon energy.  This is the x-ray photon energy that would produce an optimized image with respect to contrast and dose.

The ideal or "perfect" x-ray spectrum for mammography would be made up of photons all having the same energy (mono-energetic) and with the ability to adjust the energy for different breast conditions.  That is represented by the vertical line shown here positioned at the optimum energy for this particular breast.

If the spectrum (photon energy) is decreased below the optimum energy there will be an increase in the dose because of the decreased penetration through the breast.

If the spectrum (photon energy) is increased above the optimum energy there will be a decrease in the contrast because of the increased penetration through the glandular and pathologic tissues.

For every breast condition there should be a photon energy that is optimum for producing the best contrast to dose relationship.

The optimum photon energy (spectrum) for mammography depends on the size and density of the breast.

With a thin compressed breast that is relatively easy to penetrate the optimum energy for the best contrast to dose relationship is relatively low as shown here.

For thicker and more dense breast there are two differences compared to the thin.

  1.A higher photon energy is required for the optimum contrast to dose relationship.

  2. The contrast to dose "ratio" will be less, primarily because of the increased dose caused by the decreased breast penetration.

This general concept of an optimum x-ray spectrum for various breast sizes was published in the 1970s in the form shown here.  The contrast to dose ratio was expressed as a Quality Number and is plotted for the range of photon energies.

The three important points to be observed are:

1. There is an optimum photon energy that depends on breast size.
2. The optimum photon energy increases with breast size.
3. The image quality (contrast) to dose relationship is significantly less with the larger breast.

If we had the ideal mammography equipment this is what it would be.  It would produce a mono-energetic (one energy) spectrum and the energy could be adjusted to achieve the optimum balance between contrast and dose.

At this time we do not have the ideal machine so must use equipment with a combination of factors to produce a spectrum that is "close" to the optimum spectrum for the range of breast conditions. 

That is what we will do now.

The x-ray spectrum is determined by a combination of three (3) factors:

1. The x-ray tube anode material.

2. The material used for x-ray beam filtration.

3. The KV.

Anode: Most mammography equipment uses molybdenum anodes. Some systems have a dual-track anode that permits the operator or automatic exposure control system to select between either the molybdenum or rhodium as discussed later.

Filter: Molybdenum is the usual filter material but some systems also have an alternative rhodium filter that can be selected.
The molybdenum filter should be used only with the molybdenum anode as explained later.  The rhodium filter can be used in combination with both the molybdenum and rhodium anodes.

Tungsten anode x-ray tubes and aluminum filtration, the standard for all other types of radiography, is not used for mammography.

Let's recall that there are two types of x-radiation produced when electrons hit the x-ray tube anode.  Bremsstrahlung is the most common and is in the form of a broad continuous photon energy spectrum with a maximum energy determined by the selected KV value, that is 26kV as illustrated here.  Characteristic radiation is produced under certain conditions and is confined to just a few  photon energies represented here by the two vertical lines. 

The photon energies of the characteristic radiation is determined by the atomic characteristics of the anode material.  It varies with the atomic number (Z) of the material.

Molybdenum, and in some cases, rhodium, are materials that produce characteristic x-radiation that is near the optimum energy for mammography.  That is why they are used for the anodes.

In addition to the desired characteristic radiation near the optimum imaging energy, a continuous spectrum is produced extending up to the set KV value. This is generally an undesirable part of the spectrum because it is more penetrating than the radiation near the optimum energy and has the effect of reducing contrast.

The filters used in mammography are based on the "k edge" principle and attenuate or block the radiation above the k-edge energy of the specific filter material, either molybdenum or rhodium as illustrated in more detail later. 

Here we see the molybdenum spectrum in more detail showing both the characteristic radiation and the bremsstrahlung.

Because of the size (atomic number, Z) of the molybdenum atom the characteristic radiation is produced at two energies, 17.9keV and 19.5keV as shown here.  This is close to the optimum energy, especially for smaller breast without extensive dense tissue.

As pointed out previously the bremsstrahlung spectrum extends on up to the selected KV value, 25kV in this example, and generally reduces contrast and needs to be removed with a filter.

An interesting question.... why is the KV set to values ranging up to 30kV or 32kV if this produces undesirable radiation?

Answer...in addition to controlling the high-energy part of the spectrum and the maximum energy, the KV also has an effect on the efficiency of x-ray production.  That is the amount of exposure produced for a specific MAS value.  This higher x-ray tube output is needed for the thicker and more dense breast in order to produce an image in a reasonably short exposure time.  Also, increasing the KV does shift the spectrum, even below the filter cut-off energy as described below, so that it is somewhat more penetrating and appropriate for larger breasts.

The next step to produce an optimized spectrum is to use a filter to attenuate or block that part of the bremsstrahlung that is above the desired energy range. That is achieved with filters based on the "k edge" principle.

A quick physics review... Recall that photoelectric interactions occur when the energy of the x-ray photons exceeds the binding energy of the electrons in the material.  This is the so-called k-edge when we are considering the attenuation produced by the k-shell electrons in the atoms.  Think of the k-edge as a boundary between photon energies that are not attenuated (below the k-edge energy) and the photon energies that will be attenuated (above the k-edge energy).

The highest rate (coefficient value) occurs for photons that have energies that are just above the k-edge energy of the filter material.  The k-edge energy is equivalent to the binding energy of the k-shell electrons which is determined by the size (atomic number, Z) of the atom.  In principle, a filter material can be selected to position the k-edge and filter cut-off at any desired energy.  

Molybdenum, with an atomic number (Z) of 42 has a k-shell binding energy and its k-edge at an energy of 20.0 keV.

Rhodium, with an atomic number (Z) of 45 has a k-shell binding energy and its k-edge at an energy of 23.22 keV.

When the molybdenum filter is selected as shown here it attenuates and blocks much of the bremsstrahlung spectrum above the energy of 20 keV.  This results in the spectrum that is most often used in mammography, produced with the "moly/moly" anode/filter combination.

Many mammography systems have an alternative rhodium filter that can be selected by the operator or AEC.

With the rhodium filter the k edge boundary is shifted to a higher energy (23.22 keV) so that the portion of the bremsstrahlung between 20keV and 23,22keV is added to the x-ray beam.

This makes the beam more penetrating than when using the molybdenum filter and provides some advantage when imaging larger or more dense breast.

Rhodium is an alternative anode material that can be selected to produce a more penetrating x-ray beam than the more conventional molybdenum anode.  It is available in some mammography systems in the form of a dual-track (molybdenum and rhodium) x-ray tube anode.  The operator or AEC selects the one that is optimum for a specific patient based on breast characteristics, especially density.

Now for the physics....rhodium with an atomic number (Z) of 45 has its principal characteristic radiation at an energy of 20.3 keV with a less intense emission at 22.7 keV.  This is compared to molybdenum with an atomic number of 42 and a principal characteristic energy of 17.6 keV with the less intense peak at 19.7 keV.

The rhodium filter, with a k-edge cut-off at 23.22 keV,  is always used with the rhodium anode.  If the molybdenum filter is used, its k-edge cut-off from 20.00 keV upward would attenuate the rhodium 20.3 keV and 22.7 keV radiation.

The x-ray beam spectrum is one of the most critical factors that must be adjusted to optimize a procedure with respect to contrast sensitivity and dose.

We can think of it as a three-step procedure:

1. Select the appropriate anode (moly or rhodium)

2. Select the appropriate filter (moly or rhodium)

3. Select the appropriate KV (In the range 24 kV to 32 kV)

Increasing the KV has two effects on the x-ray beam.  It increases the efficiency and output for a specific MAS value and it shifts the photon energy spectrum upward so that the beam becomes more penetrating.

While a more penetrating beam does reduce contrast sensitivity it is necessary when imaging thicker and more dense breast.  Therefore compressed breast thickness is the principal factor that determines the optimum KV.

Mammography systems have indicators that display the thickness of the compressed breast.  This along with a general assessment of breast density is used to manually select an optimum KV either from experience or an established technique chart.

The general goal is to increase the KV as necessary to keep the exposure time, MAS, and dose to the breast within reasonable limits as breast thickness increases.

The automatic selection of the KV is a design feature of some mammography systems.  This is often based on a short, low-level, "pre-exposure" that is used to measure the penetration characteristics of the breast.  From this, a KV value is calculated and automatically set for the procedure.  Note:  this is combined with the automatic selection of anode and filter combinations for equipment that has that capability.

The automatic selection of technique factors is a valuable function because it is based on actual measurements of breast characteristics.  However, it should be used by experienced radiographers who monitor the selected protocols and image quality.

Automatic systems provide an opportunity for the operator to have some control by setting a "priority" that will shift the balance between contrast sensitivity and reduced dose.  This function is illustrated here using a 4 cm compressed breast.  If the "contrast" priority or mode is selected the automatic system would select a relatively low KV value (for example, 25 keV).  If the "dose" mode is set a higher KV value will be selected (for example, 29 keV).

A word of caution....there might be the temptation to select the "dose" mode with the expectation that it will reduce the dose to the patient.  It will, but at the cost of reducing contrast sensitivity which might have an adverse effect on the visibility of some pathologic conditions.

The selection of technique factors (anode, filter, and KV) to optimize a procedure with respect to the balance between contrast sensitivity (image quality) and dose requires an educated and experienced staff.

Even though a breast is relatively small compared to most of the body, it is still the source of significant scattered radiation that reduces contrast as illustrated  This must be considered because in mammography we are attempting to see many low contrast structures where any reduction in contrast  can be detrimental.

Two methods are used to reduce the scattered radiation to the receptor in mammography.

Grids are used for most procedures.  They are especially designed for mammography with relatively low grid ratios and material between the strips that have low absorption characteristics because of the low-energy x-ray spectrum.

The usual procedure when doing magnification mode is to remove the grid and rely on the air gap to decrease the intensity of the scattered radiation.

As the x-ray beam exits the breast it contains an image in the form of different levels of exposure produced by the variation in penetration through the different densities that make up the physical contrast.

The contrast that makes up the image is represented by these different exposures reaching the receptor.  A general knowledge of the range and distribution of the exposure values is helpful in optimizing the contrast of the breast imaging process.

The histogram, as shown here, shows the distribution of exposures for an example mammogram.  While the concept of the exposure histogram can be applied to film imaging (as we will do here) its real significance and use comes in digital mammography where the digital processing to optimize the contrast is based on modifying or selectively enhancing certain segments (as in windowing) of the histogram. 

There are two very different methods for recording and displaying mammograms:  film and digital.

Each method has distinct design features that determine the contrast characteristics of the imaging process and the necessary steps to optimize the contrast sensitivity and visibility of the breast.

With film based mammography, film serves as the media for recording within the receptor, transporting and storing images, and is the image display device.  The significant characteristic is that the contrast of the image is "fixed" and cannot be changed after the film is exposed and chemically processed.  That is why a major effort must be made to optimize the exposure and processing conditions.

Digital mammography offers many advantages over film with respect to contrast as will be described later.  In addition to being a receptor with a wide dynamic range (latitude) it provides the opportunity for using digital image processing to enhance the contrast characteristics for visualizing specific features and also the ability to control and optimize the display and viewing.

We will now first explore the contrast characteristics and issues of film and then move on to digital.

Let us recall that the principle function of film is to convert x-ray exposure (after it is converted to light by the intensifying screen) into a visible film density, or darkness.  This is the process that converts the invisible x-ray image into a visible image for viewing.

The contrast within an image is determined to a great extent by how a specific film transfers exposure values into density values.
It is not a linear, straight-line, relationship but follows a "curved" relationship as we see here.  It is the shape of this curve that describes the contrast transfer of a specific film.  This is the so-called characteristic or H & D curve. The rate at which exposure contrast is transferred to visible contrast is determined by the slope or steepness of the curve at each point along the way.

The transfer of contrast is represented by the slope or steepness of the curve.  The contrast curve is a plot of the slope of the characteristic curve which shows the film contrast at each exposure value.  If we follow the curves to the higher exposure values we see that the characteristic curve begins to become less steep indicating reduced contrast.  This places an upper limit on the exposure range that can produce contrast and an acceptable image.  The range of exposure that results in good image contrast is the latitude of the film.  In digital terminology we would refer to this as the dynamic range!

The requirement for maximum transfer of contrast in mammography is that the exposure histogram fit within the range of the contrast curve, that is within the film latitude or dynamic range.

Three steps in achieving this are:

1. Designing film specific for mammography that has both a wide latitude and high contrast transfer as discussed later.

2. Reducing the range of the exposure histogram, primarily with breast compression.

3. Carefully controlling the exposure so that the histogram is properly positioned within the latitude.

The overall contrast characteristics and contrast sensitivity of the procedure  are the combined effects of four very specific factors shown here.  They are:

1. Film Design Specific for Mammography

2. Level and Range of Exposure

3. Chemical Processing

4. Viewing Conditions

Each of these factors requires attention and inclusion in a Quality Assurance and Control Program for each mammography facility.

The challenge is that there are many variables associated with film.  These variations and sub-standard conditions that might occur can significantly deteriorate the contrast sensitivity and visibility.

We will now consider each factor with an emphasis on steps to achieving the necessary image quality. 

For mammography we need two film characteristics that generally are conflicting with each other.  First, we need a steep characteristic curve because that represents high contrast transfer and contrast sensitivity.  However, for the usual range of film densities that can be viewed on a conventional viewbox, a steep characteristic curve results in a reduced latitude.  A wide latitude is required to image the rather wide range of exposure coming through the breast.  While compression  is useful in providing a more uniform breast thickness, and a smaller range of exposure, there is still a considerable range because of other variations in thickness (near the nipple) and in density.

The contrast characteristics for mammography film are different from other radiographic films in order to have both high contrast (that is a steep slope) and a wide latitude.  This is achieved by designing the film to record contrast over an exposure range that extends to the higher film densities (darkness) compared to general radiographic film.

To utilize this extended contrast characteristic to the full advantage requires two things.

The film must be exposed to a relatively high average density (darkness) so that it is centered within the film's extended sensitive range (latitude). 

This is achieved by calibrating and setting the AEC to produce a relatively high density (a density of 1.7 is illustrated here) when imaging a test device (phantom) of uniform thickness.

A point about optimization....If the average density is set to a lower value ( 1.2 illustrated here) that is more like what is used in general radiography, the dose would be reduced.  However, the contrast would be reduced.  This is an example where attention must be given to optimizing the contrast to dose relationship.  It also illustrates the point that there are times when a certain dose level is needed to achieve the necessary image quality.

The second requirement is that the properly exposed mammography film is relatively dark (high density) and must be viewed on a specially designed bright viewbox as discussed in more detail later.

 This is a good time to revisit a major reason for using good compression.

With the compressed breast and the resulting more uniform thickness, the range of exposure (exposure histogram) is reduced and hopefully will fit within the film latitude (dynamic range).

Most mammograms are made using Automatic Exposure Control (AEC).  The AEC system measures the exposure that reaches the receptor after penetrating the breast and turns the exposure off when the necessary exposure has been delivered to produce the expected film density.

While AEC is a valuable function for producing optimum film density and visibility, it does not always produce the "perfect" exposure.  There are several potential sources of error that must be considered as illustrated here. Two are associated with the set-up and calibration of the system by the engineers, and two are under the control of the technologist/radiographer.

AEC Calibration (Physicist and Engineer Function)
The AEC must be calibrated by the engineering staff to produce the desired film density.  The calibration is verified by a medical physicist who specifies a density value that is optimum for the specific clinical facility.  The calibration is specified in terms of the film density (a value of 1.6 is illustrated here) produced when imaging a test device (phantom) of uniform thickness.

AEC Tracking  (Physicist and Engineer Function)
Tracking is the ability of the AEC to maintain correct calibration over the range of receptor types, KV values,  and exposure times used in a clinical facility.  This is generally not a problem with modern mammography equipment but is  evaluated periodically by a medical physicist in the context of the QA program.

AEC Sensor Position (Radiographer Function)
Typical mammography systems have multiple radiation sensors, or at least multiple sensor positions, for measuring the exposure reaching the receptor.  The film will be exposed to the calibrated density in the anatomical area over the selected sensor or sensor position.

The appropriate sensor position should be selected by the radiographer based on breast characteristics, especially the presence of dense areas.

An incorrect sensor location can result in an exposure error (too light or too dark) to the film.

Density Control Setting (Radiographer Function)
The radiographer can adjust the "Density" control to change the film density from that produced by the AEC normal setting.  Typical settings of the Density control are: (N)ormal, +1, +2, +3, -1, -2, -3,etc.  Although there is no standard relationship, changing the Density control by one unit will generally increase or decrease the exposure about 15%.
The Density control is useful when it appears that the AEC (N)ormal setting does not produce the appropriate film density (until it can be recalibrated by the engineer) and when certain breast conditions are better visualized with lighter or darker films.

Associated with the Density control is a function for indicating which receptor (film/screen combination) is being used.
The AEC must have this information to make the correct exposure.

We recall that the formation of a visible image on film is a two-step process.  First, the film is exposed to form the invisible latent image and then the film is chemically processed to develop the visible image.

Processing is a critical step requiring special attention in mammography because of the many sources of variability and sub-standard processing.  The details of film processing and the necessary Quality Assurance procedures are covered in the reference below.  Our purpose here is to take a brief look at the factors associated with variations in mammography film processing.

Film processing is a four step process: development, fixing, washing, and drying.

It is the development step that converts the exposed film to one with density  and contrast.  Development is not an instantaneous process but occurs over a period of time (usually about 25 seconds) as the chemicals interact with the exposed silver halide crystals in the film emulsion. 

The objective is for the development process to continue until the film is fully developed, but not overdeveloped which produces one form of film fog.

The final level of development is determined by a combination of physical and chemical factors as shown here.  These are the factors and conditions that must be addressed when setting up the processing and monitoring (either directly or indirectly through sensitometer ) of it in the context of a Quality Assurance Program.

There are two very specific processing goals:

1. An appropriate level of processing so that all (ideally) of the exposed silver halide crystals are converted to black silver and film density.
   This is necessary to get both maximum contrast and the optimum exposure sensitivity or "speed" (to reduce patient exposure) from the film.

2. Consistent processing so that the film exposure sensitivity does not drift or change with time resulting in exposure errors.

Reference:  The Photographic Process and Film Sensitivity.
 The Physical Principles of Medical Imaging, Online,  http://www.sprawls.org/ppmi2/FILMSEN/

Maximum contrast and visibility is obtained when the receptor exposure, represented by the histogram, is positioned under the contrast curve (within the film latitude or dynamic range).

Errors in obtaining the optimum film exposure can occur if either the exposure technique factors are not correct for reasons discussed previously, or if there are variations in the receptor sensitivity or "speed" over time or from one patient to another.  Some of the possible sources of unwanted changes and variations in receptor sensitivity are shown here.

Variations in Intensifying Screens.  Generally, all of the screens in a facility should have the same sensitivity or "speed" so that they will not introduce exposure errors as the various cassettes are used.  A potential source of variability is when some new screens are acquired and mixed in with the older screens,  There are simple QA tests that can be performed to verify that the screens in a facility have approximately the same sensitivity.

Variations in Film Sensitivity.  The major film manufacturers have extensive quality control procedures to reduce the variations in the sensitivity of the films they provide.  However, there can still be some variations in the film received by the user that might result in exposure errors.  There are simple QA procedures (using sensitometry) to check for film consistency, especially when changing the types of film, etc.

Exposure Time Effects on Sensitivity.  A characteristic of film is that its sensitivity might change with changes in exposure time, even when the total exposure (MAS) is the same.  This is the so-called law of reciprocity failure.  The significance is that various combinations of MA and exposure time that produce the same MAS (and total exposure to the film) might not produce the same density and contrast in a film because of this shift in sensitivity over the range of exposure times times used in mammography. This comes from how the light photons exposing the film interact with the individual crystals in the film emulsion.  When there is a high intensity exposure in a short time the effect (resulting film density) might be different from a lower intensity exposure delivered over a longer time, even when the total exposure is the same.

This problem is addressed in mammography by having the AEC system adjusted and calibrated to compensate for the reciprocity failure of the film being used.

Variations in Processing.  Variations in the chemical processing are generally a major source of exposure error because the level of processing affects the film sensitivity.  That is why mammography requires an active QA program to maintain processing consistency.

The display and viewing of the film is the last step in the total process of visualizing the anatomy and pathology within the breast.

It can be a "weak link" and reduce much of the contrast sensitivity developed in the other stages of the imaging process.

There are three (3) specific factors associated with the viewing that must be addressed.

Viewbox Luminance (Brightness).  A characteristic of the human visual system is that maximum contrast sensitivity requires a relatively bright or well illuminated image for viewing.

Film mammography presents a special challenge because the films are exposed to a relatively high density, compared to other radiographs, in order to obtain the maximum contrast from the film, as described previously.

These darker or more dense films require illumination with an especially bright illuminator or viewbox in order to enhance the visual contrast sensitivity and visualization of low-contrast objects in the breast.

Viewboxes designed for mammography have brightness values of at least 3500 nits compared to  around 1500 nits for conventional radiography illuminators.

Masking.  The advantage of a bright viewbox brings a problem.  If there are uncovered areas around a film this creates a bright light shining right into the eyes of the viewer.  This is the bright headlights of oncoming traffic situation.  Bright light shining into the eye reduces contrast sensitivity and visibility of relatively low-contrast objects.
The solution is to cover or mask the areas around a film.

Room Illumination.  Low-level illumination in the film reading room or viewing area increases visual contrast sensitivity as the eyes adapt to the darker environment.

Digital mammography provides several advantages over film for optimizing the contrast transfer from the breast to the image display and the maximizing the overall contrast sensitivity.

Three (3) of the major features are shown here.

Digital Receptor Dynamic Range.  A valuable characteristic of most digital receptors is a constant sensitivity over a wide range of exposures. This is very different from the relatively narrow latitude or dynamic range of film as we have see earlier.

The advantage is that the full exposure histogram will be easily covered by the wide dynamic range and that a considerable variation in exposure to the receptor (exposure error) can be tolerated without loss of contrast.

The transfer of exposure contrast into digital image contrast is represented by a linear (straight-line) rather than the steep characteristic curve of film with its limited latitude.  The digital image recorded by the typical digital receptor will have relatively low contrast (it would look like a rather gray image) but it will be uniform over the full exposure range.  The next step is to select the exposure range representing the actual image, that is the histogram, and to enhance the contrast by digital processing and windowing.

Digital Image Processing. One of the great advantages of digital imaging is the ability to apply a variety of processing procedures to change the image characteristics, hopefully to improve quality and visibility in most cases.  Here we a focusing attention on the contrast.  Contrast processing is common in most forms of digital radiography and is used to make the digitally acquired radiographs "look like" more conventional film radiographs with respect to contrast.  This processing can be thought of as applying a film characteristic (H & D) curve as illustrated here.  The advantage is that the user can select from many different "film characteristics" to meet the needs of specific clinical procedures.  For example, in general radiography, one "characteristic curve" type would be appropriate for chest imaging while another would be used for imaging the extremities.

In digital mammography the various contrast processing procedures are generally built into the system and might vary to some extent from one manufacturer to another.

Windowing.  Windowing,  as used in the display and viewing of most digital images (including CT, MRI, etc) is the last step in optimizing the contrast and visibility of specific objects and structures within an image.

In summary....the various contrast characteristics of digital imaging (wide dynamic range, processing, and windowing) can be combined to produce maximum contrast sensitivity as required in mammography.

Blurring is one of the five basic image characteristics that determines overall image quality.

Some amount of blurring occurs in all imaging procedures and the major clinical effect is that it reduces the visibility of detail (small objects and structures).  Blurring is especially significant in mammography because of the need to image very small calcifications.

Of all medical imaging procedures, mammography is the one designed to produce images with the least amount of blurring and the greatest visibility of small objects.

There are three (3) observable effects of blur that we will review here.

Unsharpness. Blurring reduces the perceived "sharpness" of an image or picture.  The term "unsharpness" is often used as an alternate name for blur.  However, what we see as unsharpness is an effect that is caused by the more fundamental process of blurring that occurs during the formation of an image.

Spatial Resolution. Spatial resolution describes the ability of an imaging procedure to produce images in which closely spaced objects can be seen as separate objects or "resolved".  This resolving ability is reduced by the blurring that in effect, blurs the objects together. 

Resolution plays a role in medical imaging because it can be measured by imaging certain test objects or "line-pair phantoms".

Measuring resolution is an indirect method of measuring the blurring in an imaging procedure and is useful for evaluating the performance of imaging systems.

Visibility of Detail. The clinically significant effect of blurring is that it reduces the visibility of small objects or detail in an image.

The visibility of detail (effect of blurring) is routinely measured in mammography by imaging the accreditation phantom introduced earlier and shown again here.

It contains clusters of simulated micro-calcifications varying in size.  These are the star-shaped clusters.

It is generally expected to see four of the clusters, including the ones that are 0.24 mm in diameter.

This is a general illustration of the ability of mammography to produce images of very small objects.  This is not possible with other x-ray procedures.

The blurring that limits visibility of detail comes from three (3) sources:

1. Motion (if it is present)

2. The Focal Spot

3. The Receptor and Display System.

Hopefully motion is eliminated by good compression and keeping exposure times as short as possible.

Let's now look at the other two and how they produce a combined effect.
 

The blurring produced by the focal spot depends on two factors:

1. The effective size of the focal spot.

2. The location of the object (think calcification) in the space between the receptor and the focal spot.

The amount of blurring expressed at the location of the object (and therefore relative to its size) is the product of these two factors when the location is expressed on the S scale as shown.

Focal spot blurring is reduced by using small focal spots and keeping the breast as close to the receptor as possible.

All mammography receptors, both film/screen and digital, introduce some blurring into the imaging process and are designed to keep it to an acceptable level.

The major source of blurring is the spreading of light (or electrons) within the thickness of the x-ray absorbing layer in the receptor.

The film/screen combination used in mammography consists of a single thin screen used with only one film as shown here.

The dimensions of the blur produced within a receptor is generally related to thickness and is a tradeoff with x-ray absorption and patient exposure.

The relatively thin screens used in mammography produce significantly less blur than the receptors for other radiographic procedures.

Receptors for digital mammography are designed for approximately the same amount of blurring by using a combination of thin x-ray absorbers and small pixel.

We recall that small pixels are produced by using a large image matrix size.

The significance of blurring, and its effect on visibility, depends on the amount of blurring at the location of the object (think calcification again) within the space between the receptor and focal spot as shown here.

Important point...receptor blurring is reduced (relative to the size of an object) by moving the object away from the receptor.

In other words, introducing magnification reduces receptor blurring.

The same geometry applies to receptor blurring as to focal spot blurring, except that the direction is reversed.

Some amount of blurring from both the focal spot and the receptor is present in all mammograms.

The contribution from each of these sources depends on the location of the breast along the S scale in the space between the receptor and focal spot.

Now for "mathematical magic" (we will not go into the equations), for every combination of focal spot size and receptor blur value, there is an object location where the combined or composite blur has a minimum value.

The location of that point of minimum blur depends on the size of the focal spot compared to the receptor blur value.

Let's now see how it works.

In conventional mammography the larger of the two focal spots in the tube is used.

The example shown here is an effective focal spot size of 0.45 mm.  While this is a relatively small size compared to what is used for other x-ray procedures, it is large enough to provide the heat capacity.

Because the focal spot is significantly larger than the receptor blur value, the region of minimum blurring and best visibility of detail (calcifications) is very near the receptor.

If magnification were to be introduced with this size focal spot, the image would become more blurry and the calcifications would probably become invisible.

Geometric magnification is the technique that produces the least amount of blurring and the very best visibility of calcifications.

The requirement for effective magnification is the use of the small focal spot as shown here.

Magnification mammography, as illustrated here, is the medical imaging procedure that has the least amount of blurring and provides the very best visibility of small objects.

Question for thought...if a small focal spot reduces blurring, why don't we use it for all of the mammograms? Answer:  it does not have the heat capacity to be appropriate for routine imaging. 

It is difficult to make a precise determination of the radiation dose to a breast during mammography because of the variations in breast anatomy that are encountered and not being able to insert measuring devices, dosimeters, into the breast.

The usual procedure is to make measurements of the exposure to the surface of the breast (a value of 1200mR is shown here) and then use published tables of dose factors to calculate a quantity that is defined as the Mean Glandular Dose (MGD).

The determination of MGD values for a standard reference breast is part of the a general quality assurance and procedure evaluation program.

The objective in mammography is not to adjust the equipment and imaging techniques to produce the lowest possible dose (MGD).  It is to use imaging conditions that produce the necessary image quality (primarily contrast sensitivity and visibility of detail) without the use of unnecessary exposure to the patient.

This is the Conclusion of this module.

To return to the beginning

CLICK HERE