Review of Radiological Imaging:Mammography (Lecture 007)
The content of following review is designed by and copyright to Dr. Walter Huda and Dr. Richard Slone
The online-quiz is technically designed and programmed by Dr. Jun Ni at Dept. of Radiology, University of Iowa
I. Diagnosing Breast Cancer
- Breast cancer accounts for 32% of cancer incidence and 18% of cancer deaths in women in the United States.
- The National Cancer Institute estimated that there were approximately 192,200 new cases of breast cancer in the United States in 200l.
- The number of annual breast cancer deaths was once 40,200 and has declined because of early detection and improved treatment.
- Breast cancer ultimately develops in one in eight women in the United States.
- Fig. 7.1 shows breast cancer incidence and mortality rates.
- Early detection with screening mammography significantly reduces breast cancer mortality rates for women over 50 years of age.
- Screening asymptomatic women between the ages of 40 and 50 is controversial.
- The American Medical Association, American Cancer Society, and American College of Radiology (ACR) all recommend screening of asymptomatic women
- The ACR recommends a baseline mammogram by age 40, biannual examinations between ages 40 and 50, and yearly examinations after age 50.
Cancer detection task
- Detection of breast cancer requires specialized imaging equipment and diagnostic expertise.
- Recognition of breast cancer depends on detection of subtle architectural distortion, masses near normal breast tissue density, skin thickening, and microcalcifications.
- Microcalcifications are specks of calcium hydroxyapatite (Ca5[PO4]3OH), which may have diameters as small as 0.1 mm (100 mm).
- The small differences in attenuation of x-rays between normal and malignant tissue result in low subject contrast and make cancer detection difficult.
- Detection of microcalcifications is difficult because their small dimensions also result in low subject contrast.
- Table 7.1 summarizes the key physical properties of the major breast tissues and pathologic conditions.
- Mammography is a low-cost and low-dose procedure that can detect early stage breast cancer.
- Screen/film mammography is technically demanding and requires radiographs with excellent resolution and contrast.
- This imaging equipment must be shown to be functioning properly by means of a comprehensive quality control (QC) program.
- Dedicated mammography equipment is essential for quality and low-dose screen/film imaging.
- Modem mammography equipment uses small focal spots, low tube voltages techniques, low-ratio grids, and phototiming.
- Low tube voltages are used to maximize the relative contribution of the photolations electric effect, thereby increasing subject contrast and minimizing scatter.
- Special screens, films, and dedicated film processing are also important in mammography.
- Breast compression devices are used, and the imaging chain is configured for optimal patient positioning.
- Typical x-ray imaging specifications for dedicated mammography equipment are given in Table 7.2.
- Screening mammography normally includes a craniocaudal and a mediolateral oblique view of each breast.
- Diagnostic mammographic examinations may include additional views and magnification to resolve ambiguous findings.
II. Mammography Imaging Chain
- In visualizing breast tissue, the x-ray energy level that optimizes subject contrast is approximately 20 keV.
- Higher energy photons decrease subject contrast.
- Lower-energy photons have inadequate breast penetration and substantially increase the patient radiation dose.
- Both molybdenum and rhodium are used as target materials in the anode because they produce characteristic radiation at optimal energy levels.
- Molybdenum has characteristic x-rays of 17.9 and 19.5 keV.
- Rhodium has characteristic x-rays of 20.2 and 22.7 keV.
- For these characteristic x-rays to be produced, the x-ray tube peak voltage must be higher than these values, that is, typically 25 to 34 kVp.
- Molybdenum filters (30 f..lm thick) remove most bremsstrahlung radiation above the molybdenum K-edge energy of 20 keV, because the photoelectric absorption is high.
- Removal of this high-energy bremsstrahlung radiation improves subject contrast.
- The molybdenum also filters out the very low energy x-rays that would only contribute to patient dose.
- For rhodium anodes (Z = 45; K-edge = 23.2 keV). a rhodium filter is used to remove the high-energy bremsstrahlung radiation
- The slightly higher energy x-rays from rhodium provide better penetration of thick or dense breasts.
- Three-phase or high-frequency generators are used to minimize voltage fluctuations and reduce exposure times.
- Typical x-ray tube currents are 80 to 200 mA.
- Exposure times are usually about I second but can be as long as 4 seconds for dense thick breasts.
- The x-ray tube is tilted by about 25 degrees to minimize the effective focal spot size.
- The normal focal spot is only 0.3 mm and is kept small to minimize focal spot blur.
- The small focal spot (0.1 mm) is used for magnification mammography.
- The small focal spot can only tolerate low currents (25 mA), which can result in very long exposure times of several seconds.
- A beryllium (Z = 4) x-ray tube window is used to minimize x-ray beam attenuation.
- The heel effect (higher x-ray intensity on the cathode side) is used to increase the intensity of radiation near the chest wall, where greater penetration is needed.
- This is accomplished by placing the cathode side of the tube toward the patient.
- For a normal compressed breast (4.5 cm), a typical x-ray tube voltage is 25 kVp, and tube current exposure time product is 120 mAs on a screen/film system.
- Scatter to primary ratios in mammography range from 0.6 to 1.0.
- Although these ratios are low compared with those of general radiology, they can noticeably reduce image contrast.
- Grids are commonly used to maximize image quality by reducing scatter.
- Scatter increases with breast thickness and peak voltage.
- Mammography is normally performed using a moving grid.
- Carbon fiber is the preferred interspace material, because aluminum would attenuate too many of the low-energy x-rays used in mammography.
- A high transmission cellular (HTC) grid has been developed for use in mammography, which has a focused cellular pattern that reduces scatter in both directions.
- The HTC grid has a self-supporting structure, which eliminates the need for interspace material. and thus allows more primary radiation to reach the detector
- Typical values for grid line densities range from 30 to 60 lines/cm, and typical grid ratios are 4: 1 or 5: 1.
- Grids decrease scatter but increase patient dose up to three-fold.
- Grids are sometimes not used if the compressed breast is very thin.
- Rare earth intensifying screens, such as terbium-activated gadolinium oxysulfide (Gd2O2S:Tb), are used in mammography.
- Single screens are used, which may incorporate light absorbers to limit screen diffusion
and improve resolution.
- The photon absorption efficiency in mammography screens can be as high as 70% because of the use of low-energy x-ray photons.
- Single emulsion films are normally used to reduce receptor blur by eliminating crossover and parallax effects.
- The film is placed between the x-ray source and screen to reduce blur.
- X-rays are mainly absorbed at the front of the screen and should be closest to the film to minimize blur.
- A typical screen/film combination in mammography requires 0.05 to 0.2 mGy (5 to 20 mR) at the screen to generate a satisfactory film density.
- Mammography films generally have high gradients (over three) and a low film latitude.
- Limited latitude, however, is normally not a problem when there is adequate breast compression.
- The high-contrast film/screen with a narrow exposure latitude requires very good automatic exposure control (AEC) to censure consistent film quality (density).
- Mammography films have relatively thick single emulsions, which makes them much more sensitive to processor artifacts.
- Optimal film processing is critical to ensure high image quality, and dedicated mammography processors are recommended.
- Optimal film densities in mammography are between 1.5 and 2.0, which is higher than that of conventional radiography.
- Higher film densities are needed in mammography because this results in the best film contrast.
- Special processors with extended cycle times of 3 minutes and higher developer temperatures can be used.
- The extended development time optimizes development of the latent image, resulting in increased film speed and contrast.
- Optimal film processing requires careful QC, which results in improved image quality and reduced patient dose.
III. Clinical Imaging
- Optimal mammography requires the use of breast compression.
- Compression results in greater sharpness, less scatter, and reduced patient dose.
- Compression reduces the thickness of the breast and allows low voltages to be used, thereby improving subject contrast.
- Compressed breasts are normally 3 to 8 cm thick.
- Compression immobilizes the breast and minimizes any motion.
- Compression spreads the breast tissue, making lesions easier to detect.
- Compression brings the breast closer to the image plane (object to film distance), minimizes image magnification, and reduces focal spot blur (geometric unsharpness).
- Compression also reduces exposure times, minimizing patient motion blur associated with long exposure times.
- Spot compression may be used to achieve maximum compression in a limited region of interest.
- Compression is achieved using radio translucent paddles that have an x-ray transmission of about 80% at 30 kVp.
- Compression force is normally between 111 and 200 N (25 and 45 lb).
- The principal drawback of compression is patient discomfort.
- Magnification mammography improves visualization of mass margins and fine calcifications.
- Magnification is achieved by moving the breast away from the film using a 15 to 30 cm standoff, and by keeping the source to image receptor distance constant. 1S
- The geometric principles of magnification are illustrated in Fig. 7.3
- The magnification is the ratio of the source-to-image receptor distance (SID) to the source-to-object distance (SOD); magnification is given as SID/SOD.
- A typical SID is 65 cm, and SOD in magnification is 35 cm, so that magnification is normally 1.85.
- The presence of an air gap reduces the amount of scatter reaching the film and elim11inates the need for a grid.
- The amount of breast coverage in a single magnification radiograph is reduced.
- Small focal spots (0.1 mm in diameter) are essential to minimize geometric unsharpness.
- Use of a small focal spot requires longer exposure times, which may result in increased patient motion and blur.
- Magnification generally improves image quality.
- Film viewing conditions are very important in mammography.
- High-luminance viewboxes, low ambient light, and complete film masking should be implemented in both the radiologists and technologist's area.
- For optimal viewing of the images, bright viewboxes with luminance values of approximately 3,000 candelas per square meter (cd/m2) should be used.
- Conventional viewboxes in radiology are approximately 1,500 cd/m2.
- Viewing rooms should be darkened (below 50 lux), and hot lights should be available.
- Extraneous light decreases contrast perception.
- Regions beyond the mammogram border should be covered to reduce glare and thereby improve visibility of low-contrast lesions.
- A magnifying glass should be used to view microcalcifications.
- Stereotaxic localization has been developed to perform core needle biopsies.
- Stereotaxic localizations are best achieved using digital imaging systems, which eliminate time-consuming film processing.
- The field of view of digital systems range from 50 x 50 mm to 50 x 80 mm, with a pixel size as small as 25 mm.
- Digital systems use a charged coupled device to capture the light from the screen, via optical lenses or fiberoptic tapers.
- The resultant limiting spatial resolution in the imaging plane is 8 to 15l lp/mm.
- Two views of the breast are normally acquired (within 15 degrees of normal).
- Images of the lesion will shift by an amount that depends on the lesion depth, which permits a three-dimensional localization of the lesion.
- A biopsy needle gun is positioned and fired to capture the required tissue sample.
- Benefits of core needle over open biopsies are a short procedure time, minimal local anesthetic, reduced risk, and no residual scarring of breast tissue.
- The limitations of core needle biopsy devices include their high cost and the limited field of view of real-time images.
- Computed radiography has been used for screening mammography, but the low resolution (5 lp/mm) can limit visualization of microcalcifications.
- Full field of view digital systems have recently been introduced into clinical practice.
- A typical matrix size in digital mammography is 4 x 6 k, with a pixel size of 40 to 50 mm.
- The limiting spatial resolution of digital mammography is about 10 lp/mm.
- This is inferior to screen/film, which can reach 20 lp/mm.
- A major benefit of digital mammography is the ability of image processing to improve lesion visibility in underexposed or overexposed regions.
- Current clinical trials of digital mammography have shown this modality to be at least as good as screen/film mammography.
- Digital mammography is very expensive, which is a major inhibitor to rapid diffusion of this technology.
- A digital screening examination and prior examination contain 400 MB of data, which is difficult to view and manipulate using current displays.
- Digital mammograms can be processed using computer-aided diagnosis (CAD) software, which can identify malignant lesions and microcalcification clusters
- CAD systems can assign a "probability of malignancy" for each identified lesion.
- Mammography CAD software has been shown to have sensitivities as high as 90% and can identify lesions missed by mammographers.
- CAD software has also been shown to improve the performance of mammographers when used as a "second reader."
- Although CAD systems can have a high false-positive rate of up to one or two false-positives per image, these systems continue to improve and have a promising future.
III. Image Quality and Dose
- The limiting spatial resolution of state-of-the-art mammographic screen/film combinations is 15 to 20 lp/mm.
- Magnification imaging with a small focal spot can improve the achievable spatial resolution.
- Quantum mottle is the major sources of noise in screen/film mammography.
- Film granularity is a secondary source of image noise in screen/film mammography.
- Increasing voltages reduces exposure time and patient dose when film density is kept constant.
- Low-voltage techniques increase contrast but also increase patient dose (Fig. 7.4):
- Grids may improve the contrast by a factor of two, but also increase the radiation by a factor of two to three.
- Then screen/film cassettes must be meticulously cleaned and carefully handled to minimize artifacts and maintain high image quality.
- The glandular tissue in the breast is sensitive to cancer induction by radiation.
- The average glandular dose (AGD) is the preferred measure of dose in mammography and is determined using a special phantom.
- The AGD depends on x-ray beam techniques (kV and mAs), breast thickness, and composition (Fig. 7.4).
- A composition of 50% glandular tissue and 50% adipose tissue is generally assumed for dosimetry purposes.
- An average-sized (compressed) breast is taken to be 4.2 cm thick.
- Doses should be determined annually by a certified medical physicist.
- The AGD is obtained using the measured entrance skin exposure when imaging an ACR phantom that simulates a 4.2 cm breast with 50% glandularity.
- The x-ray beam voltage and the half-value layer also influence the AGD in mammography.
- The ACR recommends that the AGD for a 4.2-cm thick breast should be less than 3 mGy (300 mrad) per film for screen/film with a grid.
- If no grid is used, the AGD should be less than I mGy per film (100 mrad).
- Typical AGD values are between 1.5 and 2 mGy per film (150 to 200 mrad) for mammography with a grid.
- The principal risk after radiation exposure is the induction of breast cancer in the glandular tissue.
- Epidemiological studies of high-dose radiation-induced breast cancer include studies of atomic bomb survivors, tuberculosis patients who underwent extended fluoroscopy, and radiation therapy patients.
- Most radiation-induced breast cancers result from an AGD in the range of 1 to 20 Gy (l00 to 2,000 rad) with little data from doses below 0.5 Gy (50 rad).
- Radiation risks for women undergoing mammography are based on extrapolations of risk estimates made at high doses.
- Based on current risk estimates, exposing 1 million 45-year-old women to an AGD of 1 mGy (l00 mrad) may result in two excess breast cancer deaths.
- Although increasing the voltage reduces the AGD in mammography (Fig. 7.4), it also reduces image contrast and is generally not recommended.
- Risk versus benefits
- For a two-view screening examination, with a total AGD of 3 mGy (300 mrad), the (theoretical) radiation risk in 1 million examined women is about six.
- This mammogram radiation risk is equivalent to the risk of dying in an accident when traveling 5,000 miles by airplane or 450 miles by car.
- Screening 1 million women is expected to identify 3,000 cases of breast cancer.
- Without a screening program, the breast cancer fatality rate is about 50%.
- Screening programs are expected to reduce the fatality rate by about 40%, or to save about 600 lives.
- The benefit to risk associated with mammography screening is therefore high.
- It is also important to note that the benefits of screening have been demonstrated in epidemiological studies.
- The radiation risks at low doses of the order of a few mGy are theoretical and mainly based on extrapolations of observed effects at doses of the order of 1 Gy.
- Radiation doses in mammography are very low and should not deter any women from having a screening examination.
V Mammography Quality Standards Act
- Successful breast cancer detection requires high-quality images that optimize contrast and resolution with minimal radiation dose.
- A comprehensive mammography program requires the combined efforts of physician, technologists, and physicist.
- Factors effecting image quality include proper patient positioning, compression, image interpretation conditions, and exposure conditions.
- The U.S. Food and Drug Administration (FDA) developed the Mammography Quality Standards Act (MQSA), which requires all 10,000 mammography facilities in the United States to be certified.
- MQSA was passed in 1994, and the final rules became effective in April 1999.
- It is against federal law to practice mammography without certification by the FDA.
- To obtain certification, the facility must receive accreditation by an approved body such as the ACR.
- The ACR developed an accreditation program in 1990 to improve the quality of screen/film mammography.
- Accreditation is currently based on the five steps listed in Table 7.3.
- Mammography facilities meeting the ACR standards receive a certificate of accreditation in mammography.
- Some states (Ark., Calif., Iowa, and Tex.) have introduced their own mammography accreditation programs, which are generally very similar to that of the ACR.
- The mammographer is ultimately responsible for ensuring that QC requirements are met.
- The MQSA requires that a lead physician takes responsibility for meeting QC requirements.
- All interpreting physicians participate in the facility medical outcomes audit.
- Mammographers must follow facility procedures for corrective action when poor quality images are encountered
- Accredited physicians are required to have interpreted at least 200 mammograms in the previous 24 months.
- The mammographer is responsible for ensuring technologists have adequate training and identifying a single technologist to oversee the QC program.
- Technologists must have the time and equipment necessary to perform QC tests
- The mammographer is responsible for selecting a medical physicist to perform the annual testing.
- A qualified individual must be designated to oversee the radiation protection program.
- All records on qualifications, techniques, procedures, and so forth, must be properly maintained and updated in a mammography QC procedures manual.
- Technologists have well-defined QC responsibilities in all mammography facilities.
- MQSA requirements include processor QC on a daily basis.
- Processor QC is performed by exposing and developing sensitometry strips and by measuring speed, contrast, and base plus fog levels.
- All major processor problems must be corrected before clinical work begins.
- Screens and darkrooms must be cleaned every day.
- Weekly tests include obtaining an image of the ACR phantom and assessment of viewbox and reading conditions.
- The x-ray imaging equipment should be inspected every month.
- Quarterly tests include a repeat analysis and analysis of fixer retention on film.
- Repeat rates are expected to be between 2% and 5%, and caused by positioning, patient motion, and overexposure or underexposure.
- Darkroom fog, screen/film contact, and compression testing is performed on a semiannual basis.
- The technologists QC program is reviewed annually by a qualified medical physicist.
- It is estimated that these activities require approximately 160 hours per year.
- The responsibilities of the medical physicist include assessing image quality and evaluating patient dose.
- Medical physicists must be adequately trained in mammography, perform at least 6 annual tests every 2 years, and receive the required continuing medical education credits.
- Imaging tests performed annually by the medical physicist are shown in Table 7.4.
- Medical physics tests must be performed when an x-ray unit or processor is installed or reassembled.
- Equipment evaluation is required after replacement of the x-ray tube, filter, collimator, or AEC.
- Phantom images are used to assess film optical density, contrast, uniformity, and image quality produced by the imaging system and film processing.
- Phantoms are equivalent to a compressed breast (4.2 cm) with equal glandular and adipose components.
- The ACR phantom contains various sized fibers (six), speck groups (five), and masses (five).
- To pass, the phantom image must show a minimum of four fibers, three speck groups, and three masses.
- The MQSA requires more stringent equipment performance as of October 2002.
- X-ray tube output must be greater than 7 mGy/s, averaged over 3 seconds.
- Spatial resolution from focal spot blur must be no worse than I I Ip/mm parallel to the anode-cathode axis and 13 lp/mm perpendicular.
- The AEC shall maintain film optical density within 0.15 of the mean density.
VI. Alternative Breast Imaging
Ultrasound breast imaging
- High-resolution and high-frequency transducers (7.5 or 10 MHz) are used for ultrasound imaging of the breast.
- Ultrasound can improve diagnostic accuracy and decrease the need for surgical biopsy in women who have suspicious findings at screen/film mammography.
- The main clinical role of ultrasound is to differentiate cysts from solid masses.
- Ultrasound may also be used to evaluate palpable masses not seen on mammograms and for biopsy guidance.
- Ultrasound is ineffective for routine screening of asymptomatic patients.
Magnetic resonance breast imaging
- Magnetic resonance imaging (MR) may supplement conventional imaging methods in the diagnosis of breast disease.
- MR is mainly used when a mammogram results in a problematic diagnosis of breast cancer.
- MR can also be used to assess the integrity of breast implants.
- Special breast coils are used to perform three-dimensional imaging of the breast with a typical volume matrix of 128 x 256 x 256 pixels.
- Fat-suppression techniques may be used to generate T1-weighted images.
- Breast MR normally uses gadolinium-diethylenetriaminepentaacetic acid contrast (0. 1 mmol/kg).
- Contrast-enhanced MR has a high sensitivity and is better able to identify tumor margins.
- The improved sensitivity of MR may be used to determine whether patients with presumed solitary nodules actuality have multifocal disease.
- Lack of contrast enhancement from fat and scar tissue may also be used to evaluate mammographically suspicious lesions.
- Benign lesions such as fibroadenomas are often difficult to distinguish from malignancies.
- MR can distinguish silicone from enhancing tumor.
- MR-guided biopsies cannot be performed with current commercial scanners.
- Nuclear medicine breast imaging is normality requested when the conventional mammogram is difficult to interpret (e.g., dense breast or fibrosis).
- Scintimammography uses technetium 99m-Iabeled sestamibi, which is administered
- Imaging is performed early (immediately after injection) and later (60 to 90 minutes
- Malignant lesions enhance early, whereas benign lesions enhance only on later images.
- Data acquisition is usually in planar mode (128 x 128).
- Standard single-proton emission tomography (SPECT) imaging is difficult to perform
because of activity in the heart and liver.
- Positron emission tomography (PET) imaging uses fluorine-18-labeled deoxyglucose (FDG) and the level of uptake is a direct measure of metabolic activity.
- Malignant tumors have high avidity for FDG; benign lesions have low avidity and
fibrotic processes show no uptake.
- PET is used for staging, restaging, and monitoring of effectiveness of treatment for patients with breast cancer.
- PET is the nuclear medicine modality of choice.
- Light diaphanography involves shining light through the breast and detecting its transmission using special cameras.
- Clinical diaphanography screening results have been poor.
- A major problem of diaphanography is the significant amount of scatter compared with light absorption.
- Differential absorption effects appear to be caused by increases in vascular, which results in nonspecific findings.
- Thermography involves imaging the infrared radiation emitted by tissues; the amount emitted depends on body temperature.
- Carcinomas near the breast surface may thus show up as hot spots when compared with the contralateral breast.
- The ACR deems thermography to be ineffective for detecting breast cancer, and its use for this purpose is not recommended.
Hawkeye Radiology Informatics (HawkRI)
Department of Radiology || Carver College of Medicine || The University of Iowa
Copyright © 2007-09, All rights reserved. Legal Notices