Full-Field Digital Mammography and Digital Breast ... RADIOOGIC TECHNOOGY, January/February 2017, Volume 88, Number 3 CE Directed Reading Full-Field Digital Mammography and Digital

299MRADIOLOGIC TECHNOLOGY, January/February 2017, Volume 88, Number 3

CEDirected Reading

This article is a Directed Reading. To earn continuing education credit for this article, see the instructions on Page 320M.

With the exception of skin cancer, breast cancer is the most commonly diag-nosed malignancy among

American women, and only lung cancer accounts for more deaths than breast cancer (see Table 1).1 As of January 2014, more than 3.1 million U.S. women were living with a breast cancer diagnosis.1 Mammography is the stan-dard imaging examination for breast cancer detection, reducing breast can-cer mortality by approximately 49% in women who are screened.2 In 2016, nearly 4 million mammography proce-dures were performed by 8737 facilities certified under the Mammography Quality Standards Act (MQSA).3

Three types of mammography assist radiologists in screening for and diagnosing breast cancer: film-screen systems, full-field digital mammography (FFDM), and digital breast tomosyn-thesis (DBT). The first FFDM system was approved for breast cancer screen-ing in 2000 by the U.S. Food and Drug Administration (FDA), and FFDM

eventually became the digital breast screening examination of choice. FFDM is considered a modality by the FDA and is subject to MQSA requirements. At one point, separate terms were used for digital mammography (in which early images had small fields of view and primarily consisted of spot images) and FFDM. As larger field detectors were developed, the terms digital mam-mography and FFDM became virtually synonymous4; both are types of 2-D breast imaging.5 As of July 2015, more than 96% of mammography units in use in the United States were FFDM units.3

Film-screen mammography is becoming a technique of the past, a fate similar to the early technique called xeroradiography.6 Converting from film-screen mammography to FFDM is essential today as health care transi-tions to full electronic image viewing and storage.7,8

The goal of screening is to identify cancers while they are small and local-ized, which offers women a greater range of treatment options and a better

After completing this article, the reader should be able to:Explain critical changes in American Cancer Society breast screening recommendations. Identify reasons why breast density is significant to breast cancer screening and diagnosis. Analyze results from recent clinical trials comparing the clinical performance of digital

breast tomosynthesis (DBT) with full-field digital mammography. Explain how DBT affects breast dose. Discuss the challenges associated with the exclusive use of DBT for breast cancer screening.

Innovation in breast imaging techniques is leading to better breast cancer detection. Concurrently, breast cancer screening recommendations are changing, and breast density is becoming a more significant factor in breast cancer screening. Digital mammography has become the preferred screening method, with more breast imaging centers including the use of digital breast tomosynthesis (DBT). The technology’s role in breast imaging has not been clarified fully, and several clinical trials are addressing DBT. This article presents research findings on digital mammography and DBT and explores the future role of 3-D breast imaging.

Marlene M Johnson, MEd, R.T.(R)

Full-Field Digital Mammography and Digital Breast Tomosynthesis

300M RADIOLOGIC TECHNOLOGY, January/February 2017, Volume 88, Number 3

CEDirected Reading


population-based screening programs. The ACS con-sidered a comprehensive review of 5 critical outcomes of screening mammography: breast cancer mortality, life expectancy, false-positive findings, overdiagnosis, and quality-adjusted life expectancy. Evidence has dem-onstrated that screening mammography in women aged 40 to 69 years is associated with fewer breast cancer deaths and supports efficacy of breast cancer screening for healthy women 70 years and older.11

The evidence supported strength-based recom-mendations that reflect the consensus of the guideline development group. A strong recommendation indi-cates the belief that the benefits of screening outweigh any undesirable effects of the procedure. Qualified rec-ommendations mean that although the evidence points to clear benefits of screening, the balance between ben-efit and harm is not clear; in addition, a patient’s values and preferences could affect the decision to undergo screening.11

The 2015 ACS guidelines strongly recommend that women with an average risk of breast cancer undergo regular screening mammography starting at 45 years of age (see Table 2). Physicians are encouraged to use the time that normally would have been dedicated to a clinical breast examination to instruct a woman on the importance of screening; the potential benefits, harms, and limitations of screening mammography; how to watch for breast changes; and other aspects of preventive care. The American College of Radiology (ACR) and the Society of Breast Imaging (SBI) endorse the 2015 ACS guidelines for women at average risk for breast cancer.11,12

The U.S. Preventive Services Task Force (USPSTF) expressed concern about advocating regular screening of women younger than 50 years. Based on its research findings, the number of deaths averted in younger women is lower than those prevented in older women, and the number of false-positive results is higher, lead-ing to unnecessary biopsies. According to the USPSTF, beginning screening mammography at a younger age increases overdiagnosis and overtreatment. Further, the task force said that mammograms can lead to unnecessary identification and treatment of invasive and noninvasive cancers that otherwise would not have been a threat or apparent in a woman’s lifetime.13

chance of survival.1 Mammography images can display cancerous lesions that are not palpable. Furthermore, mammograms can show cancer early, while the patient is asymptomatic.9,10 Most studies suggest that finding breast cancer at an early stage and a smaller size signifi-cantly improves a patient’s outcome.1

Typically, radiologic technologists acquire screen-ing mammograms with 2 projections of each breast: the mediolateral oblique (MLO) and the craniocau-dal (CC).5 A diagnostic mammogram is indicated when a woman is symptomatic or when the screening mammogram reveals suspicious findings.10 During diagnostic imaging for breast cancer, additional mammography projections often are acquired or supplemental imaging is performed. Ultrasonography can provide additional information about a suspicious lesion. Typically, DBT is used to distinguish tissue structures, and magnetic resonance (MR) imaging is selected to further study dense breast tissue and deter-mine the extent of disease once cancer is found.1,11

Guidelines and Practice The groups that develop mammography screening

guidelines revise them often; the revisions can cause controversy among prominent professional societies and advising organizations. In 2015, the American Cancer Society (ACS) updated its 2003 breast screen-ing guidelines for women at average risk of breast cancer, including those with no11: Personal history of breast cancer. Genetic mutations that increase breast cancer risk. History of radiation therapy to the chest during

childhood.The recommendations are based on evidence

from long-term follow-up of participants in breast cancer screening trials and studies of organized,

Table 1

2015 Breast Cancer Statistics for U.S. Women1

New cases of invasive breast cancer 231 840

New cases of in situ breast cancer 60 290

Deaths 40 290

Lifetime risk of receiving breast cancer diagnosis 1 in 8


CEDirected Reading

Johnson

Breast Density Breast density describes the tissue composition of

the breast as displayed on a mammogram.15 Breasts are made up of lobules, ducts, and fatty and connective tis-sues (see Figure 1). Lobules produce milk and often are referred to as glandular tissue. Ducts are the small vessels that carry breast milk from the lobules to the nipples. Fibrous tissue and fat make up breast size and shape and hold other tissues in place. Fibroglandular tissue is the combination of glandular and connective tissue. The relative amount of glandular and connec-tive tissue vs fatty tissue varies among women. Breasts are considered to be dense if they have more fibrous or glandular tissue than fatty tissue.15,16

Breast density cannot be felt and is apparent only with breast imaging. A radiologist most commonly determines breast density during image interpretation based on breast composition appearance on the imag-ing study. A category is identified and recorded in the report. On mammograms, areas of increased breast density appear white because glandular and connective tissue block the passage of x-rays to a higher degree than does fatty tissue.15,16

In the early days of mammography, physicians believed that most women’s breasts were made up of an equal distribution of fibroglandular and fatty

Reactions to Recommendation ChangesThe ACR and SBI have stated publicly that overdi-

agnosis claims are inflated because of the methodology that the USPSTF used when calculating overdiagnosis rates.12,14 Overdiagnosis is likely in 1% to 10% of women screened largely because of ductal carcinoma in situ (DCIS). Few if any invasive carcinomas are overdiag-nosed. Most women recalled after screening receive additional mammography or sonography. Only 1% to 2% of women have a needle biopsy as a result of a mam-mogram.12

According to one radiologist, of 1000 women screened in the United States, approximately 100 (10%) are recalled for further evaluation; this equates to the percentage of women recalled after abnormal cervical (Pap) screening results.14 Of the 100 women recalled, about 56 have additional mammography projections or a sonogram that demonstrates no cancer is present. Approximately 25 have a “probably benign finding” that is followed up in 6 months. Only 1% to 2% of screened women are referred for image-guided needle biopsies. In total, 15% of lesions prove to be cancerous.14

Table 2

2015 American Cancer Society Recommendations for Women at Average Risk for Breast Cancer11

RecommendationStrength of Recommendation

Women should begin regular screening mammography at age 45 years.

Strong

Women who are aged 45 to 54 years should receive annual screening.

Qualified

Women who are 55 years and older can begin biennial screening or continue annual screening.

Qualified

Women can begin annual screening between the ages of 40 and 44 years.

Qualified

Women should continue screening mammography as long as they have good overall health and a life expectancy of 10 years or more.

Qualified

Clinical breast examination for breast cancer screening is not recommended for average-risk women at any age.

Qualified

Pectoralis muscle

Rib cage

Nipple

Lobule

Areola

Ducts

Serratus muscle

Fatty tissue

Figure 1. Anatomy of the breast. © ASRT 2010.


CEDirected Reading


breast cancer risk. In an effort to improve communica-tion regarding breast density, the law would require mammography facilities to include information regard-ing each patient’s individual measure of breast density in the written reports of mammography results provided to the patient’s physician and to the patients. The provider determines which qualitative assessment tool of breast density to use. Under the proposed law, the summary

tissue, resulting in the description of a “50/50 breast.” Later studies proved that the 50/50 breast was a false assumption.17 Today, radiologists assign 1 of 4 catego-ries of breast density when interpreting mammograms (see Figure 2). The 4 levels, along with the percentage of U.S. women who have each, are15,16: Almost entirely fatty (10%). Scattered areas of fibroglandular density (40%). Heterogeneously dense (40%). Extremely dense (10%).Breast density is influenced by genetics and affected

by age, pregnancy or menopausal status, body weight, use of alcohol, and use of certain drugs such as tamoxi-fen. Some combinations of menopausal drugs also increase breast density.15 Typically, breast density decreases with age. As women get older, fatty tissue gradually replaces fibroglandular tissue.15,16,18

In recent years, clinicians have become more inter-ested in the challenges of imaging women who have dense breast tissue. Having dense breasts is closely correlated with primary risk factors for interval cancers (those occurring between screenings or 12 months after a negative mammogram), particularly in younger women.9 The ACS reported that a woman with dense breast tissue faces a relative risk of developing breast cancer that is 4 or more times higher than risk for those without dense breast tissue.1 The reason dense breast tissue is linked to increased breast cancer risk remains under study. The risk could be from a tendency for dense tissue to mask a cancer or because tumors grow more rapidly in dense breasts.9 High breast density has not been linked to an increased risk of death among breast cancer patients.18

High breast density can cause difficulty for radiologists interpreting mammograms and is associated with reduced diagnostic accuracy. Nearly all cancers can be demon-strated in fatty breasts, but mammography sensitivity in women who have extremely dense breast tissue could be as low as 30% to 48%. An increase in breast density has been associated with an increased likelihood of false-positive results from film-screen and FFDM screening.9

The Breast Density and Mammography Reporting Act, introduced in 2015, would require mammogra-phy facilities to develop an evidence-based process for informing women of facts related to breast density and

Figure 2. The 4 breast parenchymal patterns. A. Fatty. B. Scattered. C. Heterogeneously dense. D. Extremely dense. Reprinted with per-mission from Patterson SK, Roubidoux MA. Update on new technol-ogies in digital mammography. Int J Womens Health. 2014;6:784.

A

C

B

D


CEDirected Reading

Johnson

FFDM, the time between acquisition and display is a matter of seconds.7-9,24,25

Digital images consist of picture elements, or pixels. A matrix is a 2-D collection of pixels, referred to in terms of the number of pixels in the length and width of the matrix (eg, 540 540). The maximum achievable spatial resolution is determined by the size and number of pixels. The smaller the pixel size or the larger the matrix, the higher the maximum spatial resolution. Spatial resolution is critical in breast imaging and refers to the minimum resolvable separation between high-contrast objects.7,8

Clinical Performance The development of digital mammography has been

described as the most significant change to occur in the history of mammography, creating a technological plat-form for the advancement of breast imaging techniques not possible with film-screen systems.7,8,24 In the initial years, clinical outcomes from imaging patients with digital mammography were only marginally better than imaging with film-screen systems. Likewise, trials dem-onstrated a nonsignificant change in life years gained with use of digital mammography.26 Refer to the Box for research terminology definitions.

The results of the Digital Mammographic Imaging Screening Trial (DMIST) published in 2008 created the momentum for U.S. breast imaging to shift from film-screen to digital technology.30 In other countries in which digital mammography had been in use for years, clinical trial results were demonstrating that the use of FFDM was increasing breast cancer detection rates.31 The DMIST results were significant because the study included results for 42 760 women who had received both FFDM and film-screen mammography. The authors conducted a comparison analysis of clinical performance in age groups and according to menopausal status and breast density. The results indicated that FFDM was as effective as film-screen mammography overall and that FFDM improved cancer detection in women 49 years or younger who had either extremely dense breasts or mixed breast densities, or who were perimenopausal. The only subgroup for which FFDM proved less effective in detecting breast cancer com-pared with film-screen methods included women who were 65 years or older and had fatty breasts.30

must include a statement regarding the patient’s risk of developing breast cancer based on breast density and encourage women to discuss with their physicians sup-plemental screening tests and any questions they might have on the report.19

As of fall 2016, there has been no agreement on the best method for imaging dense breast tis-sue. Supplemental imaging recommendations are unclear.10,11,20-22 As of 2016, the ACS recommends that women at moderate risk for breast cancer (ie, those with a 15%-20% lifetime risk) discuss with their physicians the possibility of adding breast MR screening to their yearly mammogram.16 However, research has demonstrated that sonography can detect more cancer than mammography alone when screening women with dense breast tissue yet increases the likelihood of false-positive results.20-22 Harms of supplemental screening with sonography or MR of women with dense breasts include higher recall and biopsy rates compared with digital mammography alone. DBT has demonstrated the ability to increase cancer detection rates in all breast density types and reduce the frequency of false-positive results, which decreases unnec-essary biopsies.20 DBT might prove useful in the future after more rigorous comparative studies of supplemental screening for women with dense breasts have been com-pleted.9,20-22

More specific research is needed to understand fully the biologic components responsible for breast density. Clinical trials also are needed on the cost-effectiveness and feasibility of supplemental imaging and best prac-tices for screening women with dense breasts.19 In the future, clinicians and recommendations might stratify women into specific screening regimens based on their breast density and its significance and challenges in imaging of breast cancer.23

Digital Mammography Mammography requires 4 functions to produce

images: acquisition, processing, display, and storage. In analog imaging, processing and display functions depend on the film-screen combination. This process is inefficient and introduces opportunities for errors at each step. In digital acquisition, processing and display of the image occur independently of one another, allow-ing potential optimization at each step. Further, with


CEDirected Reading


wider dynamic range. These tailored image processing techniques lead to easier image manipulation.7,22,26,32

The use of FFDM for screening improves clinical workf low and helps breast imaging departments or centers increase typical daily examination volume. Digital technology also makes it easier for mammgra-phers to conduct quality control procedures. Further, FFDM helps eliminate physical image storage and problems related to lost or misplaced images.24 FFDM images are easier to share and can be displayed more rapidly. Radiologists can view images and interpret and report results at the same workstation, facilitat-ing more rapid diagnosis. In addition, the radiologist can remain at the same workstation when viewing images from various breast imaging modalities.24 Teaching techniques have improved and the practice of telemammography has become possible.22 Patients have reported less anxiety with FFDM than with film-screen mammography because the examination takes less time and patients can receive a more immediate diagnosis.10,22,24

Digital mammography supports the addition of breast diagnosis technology, including DBT and computer- aided detection (CAD). CAD uses software algorithms that can recognize breast abnormalities and highlight areas on the mammogram that might need further review by a radiologist.33 CAD technology is used widely in breast cancer screening today despite limited and con-flicting data on its clinical effectiveness. In a 2013 study that involved 163 000 women aged 67 to 89 years who had a mammogram between 2001 and 2006, research-ers analyzed the mammograms acquired through either FFDM with CAD or FFDM alone. The women who had FFDM with CAD had a 17% increase in DCIS diagnosis and a slight increase in early-stage invasive cancer find-ings. The authors concluded that although CAD helped physicians find more DCIS and identify invasive cancers at an earlier stage in older women, it also led to increased use of diagnostic examinations.34

Picture Archiving and Communication Systems PACS are necessary for electronic storage of all digital

images. The storage requirements for images acquired with FFDM and FFDM-CAD are much greater than storage needed for digitized film-screen examinations. An

A second large-scale study published in 2011 by Glynn et al reported on film-screen mammography and FFDM performed from 2004 through 2009. The study included 66 479 mammograms, of which 32 600 were performed using film-screen systems and 33 879 using FFDM. Results demonstrated that the use of FFDM increased cancer detection rates and positive predictive values and decreased recall percentages.26

In nearly 15 years of research comparing the diagnostic accuracy of digital and film-screen mam-mography, digital mammograms have been shown to be equivalent to or better than film-screen mammo-grams in a general screening population. FFDM has helped improve radiologists’ ability to identify abnor-malities and interpret differences between benign and malignant lesions.22 Digital mammography is superior to film-screen mammography in screening younger women with dense breasts because on digital images the physician can selectively optimize contrast in areas of dense breast tissue. Radiologists can enhance image contrast with simple window leveling per-formed by computer functions that provide FFDM a

Box

Research Terminology25,28,29

False-positive result – a positive result (indicating possible presence of disease) from a screening test or examination that subsequently leads to a negative result for disease.

Mean glandular dose – calculation of the average dose striking fibroglandular breast tissue.

Positive predictive value – likelihood that patients with posi-tive test results are indeed positive for the disease; typically expressed in percentage.

Recall rate – percentage of patients having screening mam-mograms asked to return for further diagnostic studies.

Sensitivity – proportion of individuals tested by a criterion (eg, mammography) who have positive findings and actually have a disease (true positive).

Specificity – proportion of individuals tested by a criterion (eg, mammography) who have negative results and are disease free (true negative).

Validity – extent to which a chosen measure appropriately answers the question being addressed.


CEDirected Reading

Johnson

In the 1960s and 1970s, medical physicists developed the core science of breast dosimetry that has become today’s standard. During this era, medical physicists began to realize that the dose to the breast’s glandular tissue and lymph nodes was a primary concern vs dose to skin, bone marrow, and adipose tissue.42 In the late 1980s, physicists developed a unit specific to mammog-raphy, the mean glandular dose, which is based on use of an approved phantom simulating 50% adipose and 50% glandular tissue.42,43,47,48 Actual patient dose depends on many factors, such as features of the equipment used; acquisition technique factors; and the composition, thickness, shape, and lateral dimension of the patient’s breasts. Generally, dose increases with larger breast size and higher breast density.42,43,47-49

The complexity of breast tissue and imaging have led medical physicists to develop sophisticated techniques for measuring radiation dose to the breast. Estimates of average glandular dose for various sizes and com-positions of breasts can be formulated using Monte Carlo simulations in place of a breast phantom.44-47,49-54 Monte Carlo techniques offer an alternative method to estimate the radiation dose realistically. The simula-tions are based on different breast compositions (eg, 0% glandular, 50% glandular, and 100% glandular) and the amount of photon transmission through the phan-tom.42,44,47,52 The simulation quantifies the amount of energy deposited in the glandular portion of the breast for various combinations of adipose and glandular tissue. The Monte Carlo method simulates dose for various types of breast composition; it would be nearly impossible to take into account the infinite number of tissue distributions that occur normally in human breasts.44,45,52 Although the Monte Carlo method could overestimate actual patient glandular dose, it provides a way to compare the dosimetry of various imaging tech-niques and protocols.44,45,47,50,52

The use of Monte Carlo simulations in medical imaging research is widespread because the technique is f lexible and helps quantify a factor (ie, dose) that is difficult to measure precisely. The metric reported in Monte Carlo dosimetric studies related to mam-mography involves conversion coefficients called the normalized mean glandular dose (DgN). DgN is defined as the ratio of the absorbed dose in the glandular tissue

FFDM study with CAD markings requires 3 to 10 times the amount of storage space needed for an average CT examination.24 MQSA requires that breast imaging facili-ties retain mammography images and records for at least 5 years, or not less than 10 years if no additional mammo-grams of the patient are performed at the facility.35,36

If a facility fails to retain mammograms for the length of time specified, even in the event of a PACS failure, the facility is in violation of the law.36 Many institutions do not purge mammography images because physicians might need past images for future reference.24 All breast imaging centers and depart-ments should invest in backup storage, ensure that staff is trained properly on how and when to perform data backups, and follow manufacturers’ recommenda-tions for equipment maintenance, including software upgrades.36

Radiation Dose in Digital MammographyMedical physicists, physicians, clinical personnel,

and patients should understand the dose from any medical imaging modality that uses ionizing radiation. This understanding is critical to comparing the pos-sible harm radiation could cause with the benefits the examination provides to the patient.37-41 Radiation dose concerns include dose from screening mammography, especially for women aged 40 to 49 years who typically begin having annual mammograms at age 40 years.38-41 The younger the woman, the more sensitive the breast tissue is to radiation.42-46

Still, mammographers should reassure patients that the amount of radiation exposure received from a mammogram today on any type of equipment is small and the risk of harm is minimal.17,37-40,46 Further, breast dose from mammography is highly regulated by MQSA, which requires that the mean (average) glan-dular dose for a single projection to an FDA-approved breast phantom not exceed 3 mGy. The average dose equivalent for a typical mammogram with 2 projections of each breast is about 0.4 mSv.37,47 In comparison, the average American is exposed to approximately 3 mSv of radiation a year from natural resources. A screen-ing mammogram delivers about the same amount of radiation as a woman would receive from background sources during a 7-week period.37


CEDirected Reading


difficulties associated with film-screen mammography. Initially, the clinical availability of FFDM systems was delayed several years because the FDA had not devel-oped an approval process for the modality. This delay slowed the development of equipment and progress of clinical trials.31 Many institutions had concerns about replacing film-screen equipment with FFDM systems and believed there was too little evidence to support the initial high cost of equipment and the substantial time needed to train mammographers and radiologists.24,55,56 In addition, DMIST results indicated that film-screen mammography often was more effective in women 50 years and older with fatty breasts. Because this group constitutes a large portion of the Medicare population, FFDM was considered a risky investment.30

The 2 projections provided by film-screen mammog-raphy or FFDM do not allow the radiologist to assess fully abnormalities and distinguish true lesions from other breast tissue. The nature of 2-D mammography makes it difficult to demonstrate potential lesions within dense breast tissue or to identify properly 2 or more normal fea-tures only separated vertically and appearing suspicious.25 On average, the sensitivity of film-screen mammography and FFDM for breast cancer detection is about 70%. The sensitivity of mammography drops to 30% to 48% for women with very dense breasts.9 Breast density and anatomical noise, a masking effect caused by tissue super-imposition, are the 2 main reasons for false-positive results and recalls.57

False-positive findings are not unusual in breast can-cer screening. A number of factors appear to increase the likelihood of a false-positive result: If it is the patient’s first mammogram. Greater breast density. Use of premenopausal hormone therapy. Screening with digital mammography. Longer intervals between screening. Lack of comparison images from previous mam-

mograms.11 The greatest potential harm of a false-positive result is an unnecessary biopsy,58 and imaging facilities make every effort to minimize false-positive results that require the patient to be recalled.9,24,26,30

In addition to false-positive findings, a key challenge associated with FFDM is an inconclusive mammogram

of the breast divided by the free-in-air kerma at the cen-ter of the field of view and the entrance surface of the breast.43,45-47,49

Radiation EffectsTwo major clinical studies published in 2011 evalu-

ated the risk of death from the radiation received during mammography breast cancer screening.40,41 The de Gelder study explored the benefit-risk ratio of screening women before age 50.41 Estimating a mean glandular dose of 1.3 mGy per projection, biennial mammography screening between ages 50 and 74 years was predicted to induce 1.6 breast cancer deaths per 100 000 women as compared with 1121 avoided deaths due to breast cancer. Lowering the age limit for biennial screening to include women ages 40 to 74 years was predicted to induce 3.7 breast cancer deaths per 100 000 women and prevent 1302 deaths from breast cancer.41 The second study by Yaffe and Mainprize assessed the risk of radiation-induced breast cancer caused by mammography screening.40 Estimates were made of the potential number of breast cancers, cancer deaths, and woman-years of life lost because of radiation exposure received in a variety of mammography screening sce-narios. For a population of 100 000 women, each receiving a dose of 3.7 mGy to both breasts, screened annually from age 40 to 55 years and biennially age 55 to 74 years, researchers predicted 86 cancers would be induced and 11 deaths due to radiation-induced breast cancer.40 Both studies concluded that the benefits of mammography screening between the ages of 40 and 74 years outweigh the risk of radiation-induced breast cancer, and the risk is small compared with the expected mortality reduction provided by mammography screening.40,41

The DMIST researchers compared the radiation dose delivered by film-screen mammography and FFDM. The mean glandular dose from digital systems was 22% lower than the dose delivered by film-screen mammography for each projection acquired.30 Most other studies analyzing FFDM dose have demonstrated that FFDM delivers a lower radiation dose to the breast than film-screen mammography.30,32,41,45,47

Challenges of Digital Mammography Digital mammography has introduced new chal-

lenges to breast imaging and has not alleviated all the


CEDirected Reading

Johnson

range of x-ray tube motion, number of views acquired, and scan duration time. Resolution, noise, artifact level, dose, tissue volume, examination time, and scan param-eters depend on the structural geometry of the DBT system.49,57,64,68

Tomosynthesis acquisition generally operates in 1 of 2 methods: the sweep mode or the step-and-shoot tech-nique. With the sweep mode, the x-ray tube moves in a continuous manner across the breast and is pulsed at the frame rate of the detector. In the step-and-shoot meth-od, the x-ray tube moves to the next position between 2 acquisitions, stops, and then transmits the angled x-ray beam while the tube is stationary. The tube might move in an arc around a point within the breast or sweep across a linear path. Some detectors are stationary and others move simultaneously with the tube.

With the moving detector system, the tube and receptor typically continue in a synchronous pattern, but the receptor is not firmly connected to the tube. This type of system relies on isocentric C-arm geom-etry in which the source and detector both rotate, or

Visit asrt.org/as.rt?DPhn8D for a diagram that displays how digital breast tomosynthesis system geometry varies.

that leads to a recall. The use of digital mammog-raphy has increased recall rates compared with routine use of film-screen imaging.22-24,30 Recalls are problematic because they increase a patient’s anxi-ety, increase patient dose, and have a negative effect on the department’s workflow. Follow-up studies add to a facility’s imaging volume and the amount of time a radiologist dedicates to the entire study.9

At the same time, recalls improve breast can-cer detection, and a balance is needed between increased sensitivity and improved specificity.9,12,58 Women who have been recalled are more likely to return for regular screening, which can reduce the number of false-positive findings.11,12 The median recall rate in the United States is approximately 10%, and U.S. recall rates are about 2 times higher than in Europe.9 Most women who are recalled for additional imaging do not have cancer.56-58

Digital Breast TomosynthesisFollowing conversion to FFDM, breast imaging

centers faced a new question: whether to complete the costly and time-consuming conversion to DBT. Tomosynthesis primarily was developed to overcome the limitations of 2-D imaging.9,14,26,59-64 The cost of DBT equipment is substantially greater than the cost of FFDM systems, and facilities must absorb additional expenses related to the increase in data storage.9 The FDA approved the first DBT equipment in 2011. The agency approved DBT as a diagnostic technique or as a breast cancer screening method when FFDM images are acquired along with the DBT study.5,65,66

Tomosynthesis Systems In tomosynthesis, a designated plane of the body,

referred to as a slice, is displayed in focus while motion blurs the anatomy above and below the plane. The x-ray tube generally is mounted above the patient and moves in one direction while the image receptor moves in the opposite direction (see Figure 3). The development of DBT detectors, along with video game technology that runs on graphic processing units, made tomosynthesis for breast imaging a clinical possibility.62,64,67,68

The structure and function of DBT systems vary among manufacturers. The systems differ in angular

Figure 3. Diagram of the digital breast tomosynthesis (DBT) system. Reprinted with permission from Patterson SK, Roubidoux MA. Int J Womens Health. 2014;6:781-788. doi:10.2147/IJWH.S49332.


CEDirected Reading


continuous loop. The size of DBT data is problematic compared with FFDM. Data from DBT studies require more time and electronic resources for reconstruction, viewing, and storage of images. Mammographers gener-ate hundreds of images of each breast on thousands of women every year, resulting in huge amounts of data that must be stored for extended periods.9 Depending on the pixel matrix of the detector, each projection cre-ates approximately 20 MB of raw data. On average, a compressed breast thickness of 50 mm results in 50 slices of anatomy and 600 MB of data. Saving the raw and reconstructed data from each projection could equate to 900 MB of data for a complete DBT mam-mogram and require an estimated 3600 MB of storage. The specific data requirements are influenced by tech-nical factors, such as detector and pixel size, number of projections, and slice interval.8,10,24,49,57,62,64

Image reconstruction for DBT relies on mathemati-cal algorithms. An algorithm is a formula or process used to solve a problem and is required for image recon-struction and postprocessing. Each algorithm affects specific image characteristics; image quality depends on factors, such as noise, image contrast, and comput-ing speed.14,49,64

Two main algorithm classifications are used for tomosynthesis reconstruction: analytical and itera-tive. Analytical algorithms—widely used in computed tomography—consist of reconstruction filters, called filtered back projection, that compensate for scanning over a limited angular range and minimize artifacts. Iterative reconstruction can affect image noise and computing time. Two examples of iterative recon-struction techniques are maximum likelihood and simultaneous algebraic reconstruction. Modified algo-rithms combine algebraic and analytical techniques and primarily are used to keep blurring and other artifacts minimized. Examples of modified algorithms are itera-tive deblurring and nonlinear back projection.49,57,64

Clinical Applications of DBTAs of December 2016, the FDA had not yet devel-

oped accreditation standards for 3-D tomography, which has created a situation similar to the introduction of FFDM to the market.31 The FDA has implemented a strict premarket approval process for each individual

partially rotate, around a common point. In DBT slot scanner technology, the center of rotation lies below the detector, and the system is coordinated in a linear man-ner instead of an arc.9,14,49,51,64,68

During DBT acquisition, the mammographer usually positions the breast close to the detector. Breast com-pression remains essential to maximize image quality, decrease patient motion, and minimize radiation dose. High frame rates and fast readouts minimize scan time and decrease the total breast compression time. The amount of compression used is similar to or slightly less than FFDM and depends on the scan time, which is based on the number of projections for a given detector frame. DBT systems generally acquire between 7 and 30 projec-tions in a single examination. DBT detectors must have a high quantum detection efficiency at low exposure levels, along with a fast readout and a high frame rate. Each pro-jection delivers a portion of the total dose.9,14,44,49,57,64,68

Most DBT systems have direct-conversion detec-tors made of amorphous selenium with thin-film transistor arrays made of amorphous silicon. The x-ray spectrum is the same or similar to the spectrum used in FFDM. The tube voltage generally depends on the thickness of the compressed breast and can be changed slightly. Mammographers can increase tube voltage and use filtration to decrease breast dose. Common anode/filter combinations include tungsten/rhodium and tungsten/aluminum.44,49,62,64

The tomosynthesis angle is defined as the angular range over which the x-ray tube moves relative to the pivot point. DBT systems have angles between 11° and 60° and all of these angles produce images with accept-able resolution. In general, the larger the tube motion, the more projections and slices produced.49,57 As the tomo-synthesis angle increases, the better the depth resolution and the smaller the effective slice thickness. Larger angles result in less image blurring.49 It is difficult to pinpoint the optimum tomographic angle for DBT because depth resolution is not the only parameter affected by the angle. A larger tomosynthesis angle prolongs scan time and can increase the risk of motion artifacts.49,57,64

Data ProcessingOnce the DBT system acquires volumetric 3-D

data, a radiologist can view images individually or as a


CEDirected Reading

Johnson

images is applied most often in countries other than the United States.77-80

Clinical Performance Trials on DBTThe clinical use of DBT has gained acceptance and

DBT research has increased.57 Clinical trials and stud-ies have investigated the advantages and disadvantages of DBT for breast cancer screening, cancer detection rates, microcalcification display, recalls, false-positive rates, breast dose, and physician interpretation times. DBT offers the advantages of FFDM, including fewer artifacts, consistent quality, and digital image process-ing.9

Diagnostic Accuracy and Cancer Detection The malignant features of breast cancers can be

classified as mass, focal asymmetry, architectural dis-tortion, and microcalcification. DBT offers potential advantages for evaluating masses, areas of architec-tural distortion, and diagnosing asymmetries over conventional 2-D mammograms. DBT can discern microcalcification clusters effectively from an under-lying mass and has proved to be more accurate at determining the true size of tumors.77-85

From 2011 to 2014, the FDA approved only 2 DBT systems for market.66 However, several DBT systems have been available since 2008 in other countries.77,78 The first 2 large breast cancer screening trials evaluat-ing the performance of FFDM-DBT were conducted in Europe. These trials were the Oslo Tomosynthesis Screening Trial, referred to as the Oslo or OTST study, and the Screening with Tomosynthesis or Mammography Trial (STORM).77,78

The preliminary results of the Oslo trial, published in 2013, demonstrated that cancer often overlooked or not visible on FFDM can be seen on DBT. During the study, FFDM and DBT images were collected on 12 631 women aged 50 to 69 years. The readers interpreted the FFDM images first, followed by interpretation of the FFDM-DBT image set. The use of FFDM-DBT compared with FFDM alone increased cancer detection by 27%, from 6.1 per 1000 to 8 per 1000, and decreased false-positive rates from 6.1% to 5.3%. Among the women studied, 121 had cancers, 77 of which were found with FFDM alone. A total of 101 cancers were detected with FFDM-DBT.

DBT system and considers DBT equipment an exten-sion of FFDM. In addition, for MQSA accreditation purposes inspectors only review 2-D images.65 Facilities that use DBT for screening must submit either the FFDM images acquired during the DBT study or the 2-D images generated from the DBT data set using an algorithm, referred to as synthesized mammogra-phy.5,65,69-71

MQSA allows certified mammography facili-ties to use accredited DBT equipment based on the unit’s 2-D imaging component. Data from 2-D digi-tal mammography is considered less effective than DBT at displaying certain abnormalities, specifically microcalcification clusters. Still, FFDM images are necessary for comparison of prior 2-D studies.71,72 As a result, DBT systems available on the market today can acquire images through a number of techniques, including FFDM alone, DBT alone, FFDM-DBT, or DBT-synthesized mammography.69-71,73-75 For facilities using a 3-D system in clinical practice, the FDA only certifies the 2-D component of the unit, which means facilities must submit FFDM images or synthesized 2-D images for review.5

As of December 2016, more than 30 FFDM and DBT systems had been approved, cleared, or accepted by the FDA since 2000. The FDA considers each manufacturer’s system a new, separate mammography modality under MQSA, and personnel operating DBT equipment must meet standard MQSA requirements for training on a new modality.76

Each manufacturer has designated indications for use and methods of image capture for breast cancer screening.5,65 For example, the Siemens Mammomat Inspiration had a premarket approved function called 2D+Tomo-Scan. When in use, the first image acquisi-tion produces a conventional 2-D image set by having the swivel arm at a 0° angle. The acquisition of 3-D data follows the 2-D acquisition.74

In contrast, Hologic’s Selenia Dimensions 3-D system with C-view software can acquire 2-D images directly from the 3-D image set, eliminating the need for FFDM images.69,75 The practice of acquiring 2 projections of each breast with FFDM for screening is the most common protocol in the United States. The double-reading practice of acquiring FFDM-DBT


CEDirected Reading


al.82 This was an important study because of the large number of women participating and the inclusion of academic and nonacademic institutions. The inves-tigators retrospectively analyzed data from 13 sites that reported results for women screened with FFDM alone and those screened with the addition of DBT. They compared screening performance measures from 281 187 FFDM images with those of 173 663 FFDM-DBT images. The study demonstrated a 41% increase in invasive cancer detection in the women who had FFDM-DBT.82

Microcalcification Detection

Microcalcifications are tiny deposits of calcium in the breast tissue frequently observed on mammog-raphy. Most calcifications are not cancerous but can indicate carcinoma in situ.9 Microcalcifications are relatively easy for radiologists to identify on FFDM images as high-contrast individual specks or clus-ters.9,89,90

The use of FFDM-DBT detected 25 additional cancers not diagnosed with FFDM. The trial’s most important finding was the 40% increase in the detection of invasive cancers that were diagnosed on FFDM-DBT images but not from FFDM alone.77

Investigators with the STORM trial in Italy released their results in April 2013. A total of 7292 women aged 54 to 63 years were invited to screen. One group was screened with FFDM alone, and the other with FFDM-DBT. Of the 59 cancers detected, 39 were found on FFDM alone and all 59 were demonstrated on FFDM-DBT images, representing a 34% increase in cancer detection.78

A similar study was conducted between 2013 and 2015 in the United Kingdom, comparing the perfor-mance standards of FFDM-DBT with FFDM alone. The tomosynthesis with digital mammography inves-tigation, or TOMMY trial, included 7060 women aged 47 to 73 years, 1158 of whom had cancer. The results demonstrated an increase in specificity with the use of FFDM-DBT compared with FFDM alone and a mar-ginal improvement in sensitivity.59

DBT is being used frequently in the United States and other countries. Supporting evidence from recent clinical trials suggests that DBT imaging of masses has demonstrated good patient acceptance, physician pref-erence for DBT images, and improvement in sensitivity and characterization of lesions (see Figure 4). The 3-D technique more clearly portrays benign findings such as lymph nodes or skin calcifications than routine FFDM and might be superior to FFDM for preoperative mea-surement of breast lesion size.9,77-88

In a small study published in 2007, Poplack et al analyzed data from 98 patients recalled after FFDM who underwent diagnostic evaluation using film-screen mammography and a DBT examination.84 The authors found that clinical image quality of DBT was either equivalent or superior to film-screen technology. In 2008, Andersson et al found that malignancies were more visible on FFDM-DBT studies than on images acquired with FFDM alone,85 and in 2013, Yang et al reported improved diagnostic accuracy for noncalcified lesions on the DBT images.83

The most significant U.S. study of DBT breast can-cer screening was published in 2014 by Friedewald et

Figure 4. Cancer is more evident with tomosynthesis. A. Cranio- caudal view of screening digital mammogram. No abnormality is evi-dent. B. Digital breast tomosynthesis slice of the same patient, in the same position, depicts a spiculated mass in central breast not evident in the screening mammogram view (arrow). Reprinted with permis-sion from Patterson SK, Roubidoux MA. Update on new technologies in digital mammography. Int J Womens Health.2014;6:784.

A B


CEDirected Reading

Johnson

Recalls Recalls refer to recommendations for further imag-

ing or workup following a screening mammogram. Approximately 25% of patients are recalled for a second mammogram following FFDM because of tissue super-imposition.92 On DBT images, overlapping tissues are separated, nearly eliminating recalls related to super-imposed tissue.14 Breast imaging centers also frequently recall women following FFDM screening because the radiologist is certain there is an abnormality, but the lesion is not evident on both mammogram projections; therefore, the physician is uncertain of the abnormal-ity’s location. With DBT imaging, the radiologist can determine the location of a lesion based on where it appears in the data set.14

In numerous clinical studies, the use of FFDM-DBT imaging decreased recall rates from as much as 30% to 40% compared with FFDM alone.14,38,39,81,82,88,93,94 In a study by Haas et al, the investigators analyzed screening performance measures for 6100 women who under-went FFDM-DBT examinations and 7058 women who underwent FFDM alone.88 The authors noted a decrease in recall rates from 12% in the women who had FFDM alone to 8.4% in women who had FFDM-DBT. After adjusting for differences in age, breast density, and risk factors, the authors reported a 38% reduction in recall rates with the use of FFDM-DBT. The recall rate reduction was significant for all breast density catego-ries with the exception of women with fatty breasts and for all age groups except women 70 years and older.88

When determining whether a woman needs a biopsy, the radiologist relies on the positive predictive value (PPV) of the mammography finding. The PPV of mammograms resulting in recommendation for biopsy ranges between 20% and 40%, with a median of 31.4%.13,14,57,84,89-91,93,94 The use of DBT imaging in breast cancer diagnosis has been shown to increase the PPV of a suspected lesion found during mammography screen-ing, particularly in young women and women with dense breasts.77-82 For example, Friedewald et al reported a sig-nificant reduction in recall rates, from 10.7% with FFDM alone to 9.1% with FFDM-DBT. The authors also report-ed a statistically significant increase in PPV for recalls from 4.3% with FFDM alone to 6.4% with FFDM-DBT.82

In an article by Rose et al, the authors analyzed recall rates from a single clinical site before and after the

The literature is limited regarding clinical micro-calcification assessment with DBT. The engineering, physics, and reconstruction of various manufacturers’ equipment can lead to different results and even oppos-ing conclusions. The x-ray source, motion, pixel size, reconstructive algorithms, and method used during interpretation all can affect the ability to see micro-calcifications on DBT images.87 Microcalcifications have slightly less contrast on DBT than on FFDM and appear individually on separate slices. Clusters can be more difficult to assess when they are not all seen at the same time and when adjacent structures have similar contrast.9,86,87,90,91

Research on DBT’s effectiveness in displaying microcalcifications has produced inconsistent results. Poplack et al reported that 14% of recalled studies involved calcifications. In 11% of the microcalcification cases, radiologists deemed the DBT images inferior to FFDM, and of all studies in which DBT was considered inferior, 72% were deemed so because of calcifica-tions.84 In another study, Sprangler et al compared FFDM images with DBT images.90 The results dem-onstrated a nonsignificant improvement in sensitivity and specificity with FFDM compared with DBT when demonstrating microcalcifications.90

Tomosynthesis can be as effective as FFDM in dem-onstrating microcalcifications when the resolution of the DBT images is maximized with small pixel sizes.14,89 Kopans et al analyzed 100 cases of microcalcifications with FFDM and DBT using small pixel sizes.89 The authors found that 92% of the cases acquired with DBT had equal or superior clarity for detecting microcalcifi-cations compared with FFDM.

A volume rendering technique also has been suggest-ed to improve diagnosis of microcalcification clusters on DBT. Radiologists should ensure that they interpret DBT images in the slab mode in addition to each slice.14,89-91 To date, the general consensus is that the industry must address the issue of microcalcification display on DBT before DBT can be considered as a stand-alone screening technique.6,57,64,84,86,90

Visit asrt.org/as.rt?3vTu34 to see how calcifications can be difficult to perceive on planes through volume but are more easily appreciated on slab images (see the article’s Figure 3).


CEDirected Reading


Different approaches for measuring DBT dose consider the unique conditions and factors of DBT acquisition.49-51,71,95-98 The air kerma for the tube output (Gy/mAs) can be estimated using 3 factors: the incident dose at the breast surface, a correction factor for breast granularity, and a factor related to the anode/filter combi-nation. To formulate a breast dose table, factors for various beam qualities or half-value layers and compressed breast thicknesses are applied and a mathematical estimate is established.44,45,52 Determining the normalized glandular dose coefficients for different beam energies, breast thick-nesses, and x-ray spectra also can be used to convert the entrance ion dose to glandular dose.37,42,50,53,54,61,95-97

MQSA standards regulate DBT dose; dose from the FFDM acquisitions acquired during the DBT study are evaluated to discern whether the system is delivering acceptable levels according to MQSA. The examination’s FFDM acquisitions must not exceed 3 mGy, the MQSA limit for a single-projection screen-ing mammography study for a CC-equivalent of an average breast. The CC projection in DBT imaging is performed in acquisition mode with the x-ray tube at the 0° position.44-46,50,52-54,87,95 One of the first U.S. clinical sites to incorporate DBT into its clinical practice found higher exposure levels for FFDM-DBT acquisitions than for FFDM alone.45 As more research followed, results consistently showed that FFDM-DBT acquisi-tions increased a patient’s breast dose in a range from 34% to 61%.37,44-46,50,52-54,95-97 In studies where FFDM-DBT systems replaced film-screen mammography without previous use of FFDM alone, the mean glandular dose for a single FFDM-DBT projection resulted in a dose higher than that of film-screen mammography.50,95,97-99

The protocol for FFDM-DBT screening used rou-tinely in the United States requires exposing a patient to radiation during FFDM and DBT acquisition; this protocol is the primary reason breast dose is significantly higher with FFDM-DBT compared with FFDM alone. The decision to examine the breast with the FFDM-DBT protocol often is left to the physician and patient. Some breast imaging centers have begun routine use of FFDM-DBT for screening. At these facilities, a woman might receive a full FFDM-DBT acquisition automatically.9,37,46

The breast radiation dose from a single DBT projec-tion is about the same as from one FFDM projection.

introduction of DBT.81 They examined a total of 13 856 FFDM screenings without DBT and 9499 screenings with DBT. The authors noted a significant reduction in recall rates, from 8.7% with FFDM alone to 5.5% with FFDM-DBT. The results also demonstrated a statisti-cally significant increase in PPV for recalls, from 4.7% with FFDM alone to 10.1% with FFDM-DBT. Among women undergoing screening for the first time, the recall rates were 13.6% with FFDM alone and 9.6% with FFDM-DBT.81

False-positive mammograms have positive results, or suspicious findings, but lead to no discovery of cancer within a year following the mammogram or recom-mended biopsy. Recalls that result in false-positive findings are problematic. In the Oslo trial, false-positive rates were reduced from 10.3% with FFDM to 8.5% with FFDM-DBT. The STORM trial demonstrated better results, with a 17.2% reduction in false-positive rates with FFDM-DBT.77,78

Breast cancer detection is lowest among women who have extremely dense breasts or heterogeneous breast density. DBT might prove to be more useful at lowering recalls related to false-positive results in this group.29,57

Radiation Dose The amount of radiation patients receive from DBT

studies depends on many factors, including37,49,50,53,54,87,95-99: Method used to measure dose. Scanning parameters of the DBT system. Breast composition. Breast size. Technique used during image acquisition.In 2014, the American Association of Physicists in

Medicine prepared a report that describes a method-ology and provides data necessary to estimate mean glandular dose from a tomosynthesis acquisition. Conventional mammography dosimetry was not designed to focus on the complex geometry of tomo-synthesis acquisition.42 In early clinical trials, DBT dose was computed with conventional dosimetry and overes-timated dose by as much as 56%.42,46 In most subsequent clinical trials, a Monte Carlo simulation technique has been used to estimate mean glandular dose of various breast sizes and thicknesses.42,46,50,53,54,61,71,95-98


CEDirected Reading

Johnson

The only way for DBT dose to equal that of FFDM is if the noise in each individual projection is determined primarily by quantum noise with minimal contribution from the electronic noise. When this is true, the individ-ual projections can be used to reconstruct an image slice that is comparable to an FFDM image. Electronic noise produced in DBT imaging cannot be eliminated entire-ly, resulting in a slightly higher dose than in FFDM imaging. Noise is the main reason high-efficiency detec-tors with minimal electronic noise are essential for DBT imaging.49,57,67

Computer-aided DetectionThe use of CAD in conjunction with DBT is mostly

experimental; a limited number of studies had been published as of summer 2016.49,100-102 The large volumes of data that would be created if CAD data were applied to each DBT image are problematic. CAD systems used for FFDM cannot be used with DBT without modification. The most common approach to develop-ing DBT-CAD systems is to create an algorithm that addresses a specific concern in breast imaging diagno-sis.49 DBT-CAD systems could prove useful to detect breast masses using projection images, reconstructed slices, or a combination of the two. Combining multiple slices into a thicker slab might prove helpful in adding CAD because the combined image would more closely resemble FFDM images.100

DBT-CAD systems also could be used to help char-acterize lesions and detect microcalcifications.49,101 A clinical study involving DBT-CAD for microcalcifica-tion detection reported a sensitivity of 85%; however, the false-positive rate was too high for clinical use.103 In the future, more CAD prototypes will need to be developed and tested if DBT is to be used consistently for screening.102

Operational Challenges of DBTWide adoption of tomosynthesis requires address-

ing some challenges of the technology, including artifacts, interpretation time, workflow issues, and training. In tomosynthesis, high-contrast objects such as surgical clips can cause substantial artifacts.57 A DBT phenomenon called undersampling also can occur. Undersampling of an object, such as a breast lesion,

Patient positioning, particularly the DBT MLO projec-tion, can cause a 5% to 13% variation in mean glandular dose.45 One way to limit breast dose in DBT is to use the synthesized mammography images from the DBT data and eliminate the FFDM exposure. Using synthesized DBT images can reduce the mean glandular dose by 45% compared with FFDM-DBT.50,70,71,73,95-98 Research results on premarket approved systems with synthesized mam-mography have demonstrated that use of synthesized images alone or in combination with tomosynthesis is comparable in performance to FFDM alone or FFDM-DBT, while successfully decreasing the dose.71,73 Eliminating FFDM acquisition improves patient com-fort because the breast is compressed for 2 projections instead of 4 projections.69 The ability to produce DBT synthesized mammography images is relatively new and not available at all clinical sites. Further, all health insur-ance plans might not cover fully the use of synthesized DBT images for breast cancer screening.37,46,71

Another approach to decreasing DBT dose is to allo-cate more (approximately 50%) of the x-ray tube output to the central projection of the DBT and distribute the remaining half over other projections.57 A third option for decreasing DBT breast dose is acquiring a single DBT projection and using DBT as a standalone tech-nique. Although the exact projection to use as the single image remains under investigation, early results indi-cate that a single projection DBT is equal or superior to FFDM studies in cancer detection and lowers total breast exposure from DBT.98

In theory, the ability to acquire DBT diagnostic qual-ity images with equal or higher quality than FFDM requires a slightly higher dose than FFDM alone. The amount of quantum noise in an image is indirectly proportional to dose. In digital imaging, the amount of quantum noise is a product of quantum mottle plus the noise that is produced from detector electronics. This type of inherent noise remains fairly constant and is influenced primarily by the efficiency of the DBT detectors. During a DBT acquisition, each individual pro-jection is delivered at a much lower dose than in FFDM because the total dose is divided among the projections. The low exposure significantly increases the amount of quantum noise in each individual projection, while the electronic noise accumulates with each projection.


CEDirected Reading


Diagnostic mammography requires additional radi-ologist time that ranges from a few minutes to several hours if the physician performs image-guided biopsy. The possibility of using only one projection from the DBT data set for interpretation also would decrease DBT interpretation time if this becomes standard clini-cal practice.14,57,103,104,106

ConclusionThe benefits of FFDM in breast cancer screening

and diagnosis have been well established in the past 15 years, and the use of FFDM has become prominent in clinical practice. The introduction of DBT imaging has initiated a new approach to digital breast imaging. Research on the effectiveness of DBT is encouraging, particularly in the areas of breast cancer detection, dense breast imaging, and reducing the frequency of false-positive results and recalls. Further technological advances continue to optimize imaging parameters. Incorporating DBT with other modalities is being studied; this includes possible contrast-enhanced tomosynthesis that creates breast images similar to MR imaging. Automated breast sonography with DBT is emerging and could prove useful in the screening of dense breasts.

Significant clinical challenges in DBT imaging include breast dose, dense breast imaging, and demon-stration of microcalcifications. Breast imaging practices also need to address the issues of equipment costs and implementation strategies to include longer interpreta-tion times, data storage, and training before converting to DBT imaging.

Data strongly supports investment in large-scale popu-lation screening trials, particularly aimed at solutions to the double acquisition of images required for 2-D and 3-D mammograms. Other research is focused on evaluating integrated synthesized and digital mammography.

Marlene M Johnson MEd, R.T.(R), has dedicated 40 years to the radiography profession as a technologist and to teaching in the radiologic sciences and developing educational programs in imaging. She spent 25 years at the University of Utah in Salt Lake City, where she completed her career after developing programs in computed tomography, magnetic

is due to the limited angular range of the x-ray tube during a DBT acquisition. Undersampling leads to out-of-plane artifacts that depend on the size and contrast of the object; the larger the structure, the more the artifact spreads out. Manufacturers have developed algorithms to reduce the repetitive appearance of out-of-plane arti-facts.49 A practical solution to this problem is the use of wide-angle tomosynthesis that results in increased angular span and lower compression.14,91

The amount of time necessary for mammographers and radiologists to learn how to perform and interpret DBT imaging must be considered when incorporating DBT into clinical practice. Training initially disrupts workflow and decreases the number of examina-tions staff can perform and interpret during a typical day.24,104,105 MQSA requires 8 hours of initial training before independently performing a DBT examination, along with additional time to learn manufacturer-specific operations. MQSA allows training to be provided by the manufacturer or a qualified instructor.24,105 Adding DBT to breast imaging also affects workflow. Breast imag-ing staff must plan for and make permanent changes to address challenges such as increased data storage and additional time for interpretation after radiologists com-plete DBT training.24

Research shows that DBT interpretation takes longer than FFDM, even after radiologists learn the modal-ity; the high number of DBT images they must review simply requires more time.24,103,104,106 Routine FFDM consists of 4 images, whereas DBT can produce 200 images or more. Interpretation time for all studies also varies among radiologists and depends on the amount of experience the radiologist has in general and with DBT images. On average, interpretation times for FFDM-DBT imaging are 47% longer than for FFDM alone.24,77,104 In the Oslo trial, interpretation time for FFDM-DBT averaged 91 seconds and FFDM required 45 seconds.77 Similar increases were found in the Dang et al study, in which FFDM-DBT interpretation required an average of 2.8 minutes per examination and FFDM required 1.9 minutes.105

Still, when accounting for the total amount of inter-pretation time required for DBT, the resulting decrease in recalls following DBT should save time that was pre-viously dedicated to follow-up diagnostic imaging.14,57


CEDirected Reading

Johnson

12. ACR and SBI continue to recommend regular mammography starting at age 40 [news release]. Washington, DC: American College of Radiology; October 20, 2015. http://www.acr.org /About-Us/Media-Center/Press-Releases/2015-Press-Releas es/20151020-ACR-SBI-Recommend-Mammography-at-Age -40. Accessed November 23, 2016.

13. U.S. Preventive Services Task Force. Understanding task force recommendations: screening for breast cancer. https:// www.uspreventiveservicestaskforce.org/Home/GetFileBy ID/2811. Accessed October 23, 2016.

14. Kopans DB. Digital breast tomosynthesis from concept to clinical care. AJR Am J Roentgenol. 2014;202(2):299-308. doi:10.2214/AJR.13.11520.

15. American Cancer Society. Breast density and your mammo-gram report. http://www.cancer.org/acs/groups/content /@editorial/documents/document/acspc-039989.pdf. Published July 2016. Accessed November 23, 2016.

16. Having dense breasts. BreastCancer.org website. http://www.breastcancer.org/risk/factors/dense_breasts. Accessed October 23, 2016.

17. Yaffe MJ, Boone JM, Packard N, et al. The myth of the 50-50 breast. Med Phys. 2009;36(12):5437-5443. doi:10.1118/1.3250863

18. Breast cancer patients with high density mammograms do not have increased risk of death. National Institutes of Health website. https://www.nih.gov/news-events/news -releases/breast-cancer-patients-high-density-mammograms -do-not-have-increased-risk-death. Published September 6, 2012. Accessed January 18, 2016.

19. Breast Density and Mammography Reporting Act of 2015, S 370, 114th Cong (2015-2016).

20. Melnikow J, Fenton JJ, Whitlock EP, et al. Supplemental screening for breast cancer in women with dense breasts: a sys-tematic review of the U.S. Preventive Services Task Force. Ann Intern Med. 2016;164(4):268-278. doi:10.7326/M15-1789.

21. Sprague BL, Stout NK, Schechter, et al. Benefits, harms, and cost-effectiveness of supplemental ultrasonography screening for women with dense breasts. Ann Intern Med. 2015;163(3):157-166. doi:10.7326/M14-0692.

22. Stout NK, Lee Sj, Schechter CB, et al. Benefits, harms, and costs for breast cancer screening after US implementation of digital imaging. J Natl Cancer Inst. 2014;106(6):dju092. doi:10.1093/jnci/dju092.

23. Hardy K. Individualized screening. Radiology Today website. http://www.radiologytoday.net/archive/rt0216p12.shtml. Published February 2016. Accessed November 23, 2016.

24. Odle T. Digital breast imaging workflow. Radiol Technol. 2016;87(3):305M-322.

resonance imaging, and a bachelor’s degree program in nuclear medicine.

Reprint requests may be mailed to the American Society of Radiologic Technologists, Publications Department, at 15000 Central Ave SE, Albuquerque, NM 87123-3909, or emailed to [email protected].

© 2017 American Society of Radiologic Technologists

References 1. American Cancer Society. Breast cancer facts & figures

2015-2016. www.cancer.org/acs/groups/content/@research /documents/acpc. Accessed January 10, 2016.

2. Nickson C, Mason K, English D, et al. Mammographic screening and breast cancer mortality: a case-control study and meta-analysis. Canc Epidemiol Biomark & Prev. 2012;21;1479. doi:10.1158/1055-9965.EPI-12-0468.

3. MQSA national statistics. U.S. Food & Drug Administration website. http://www.fda.gov/Radiation-EmittingProducts /MammographyQualityStandardsActandProgram/Facility Scorecard/ucm113858.htm. Accessed February 28, 2016.

4. Hall FM. Digital mammography versus full-field digital mammography. AJR Am J Roentgenol. 2012;198(1):240. doi:10.2214/AJR.11.7413.

5. Digital accreditation. U.S. Food & Drug Administration web-site. http://www.fda.gov/radiation-emittingproducts/mam mographyqualitystandardsactandprogram/facilitycertifica tionandinspection/ucm114148.htm. Accessed March 5, 2016.

6. Screen-film mammography: headed the way of xeromam-mography? U.S. Food & Drug Administration website. http://www.fda.gov/radiation-emittingproducts/mammo graphyqualitystandardsactandprogram/MQSAinsights. Accessed February 22, 2016.

7. Dycus P. Breast imaging basics module 10: digital mam-mography. American Society of Radiologic Technologists website. http://asrt.mycrowdwisdom.com/diweb/catalog /item/eid/105732. Accessed January 10, 2016.

8. Körner M, Weber C, Wirth S, Pfeifer KJ, Reiser MF, Treitl M. Advances in digital radiography: physical principles and system overview. Radiographics. 2007;27(3):675-686. doi:10.1148/rg.273065075.

9. Patterson SK, Roubidoux MA. Update on new technologies in digital mammography. Int J Womens Health. 2014;6:781-788. doi:10.2147/IJWH.S49332.

10. Reynolds A. Breast density and digital breast tomosynthesis. Radiol Technol. 2013;85(1):63M-83M.

11. Oeffinger KC, Fontham ET, Etzioni R, et al. Breast cancer screening for women at average risk: 2015 guide-line update from the American Cancer Society. JAMA. 2015;314(15):1599-1614. doi:10.1001/jama.2015.12783.


CEDirected Reading


.gov/radiation-emittingproducts/mammographyquality standardsactandprogram/guidance/policyguidancehelp system/ucm136287.htm. Updated August 11, 2014. Accessed November 23, 2016.

37. Understanding radiation from imaging tests. American Cancer Society website. http://www.cancer.org/treatment /understandingyourdiagnosis/examsandtestdescriptions /understanding-radiation-risk-from-imaging-tests. Revised November 30, 2015. Accessed February 20, 2016.

38. Brawley OW. Risk-based mammography screening: an effort to maximize the benefits and minimize the harms. Ann Intern Med. 2012;156(9):662-663. doi:10.7326/0003-4819 -156-9-201205010-00012.

39. Raftery J, Chorozoglou M. Possible net harms of breast cancer screening: updated modelling of Forrest report. BMJ. 2011;343:d7627. doi:10.1136/bmj.d7627.

40. Yaffe MJ, Mainprize JG. Risk of radiation-induced breast cancer from mammographic screening. Radiology. 2011;258(1):98-105. doi:10.1148/radiol.10100655.

41. de Gelder R, Draisma G, Heijnsdijk EA, de Koning HJ. Population-based mammography screening below age 50: bal-ancing radiation-induced vs prevented breast cancer deaths. Br J Cancer. 2011;104(7):1214-1220. doi:10.1038/bjc.2011.67.

42. Huda W, Nickoloff EL, Boone JM. Overview of patient dosim-etry in diagnostic radiology in the USA for the past 50 years. Med Phys. 2008;35(12):5713-5728. doi:10.1118/1.3013604.

43. Buschong SC. Radiologic Science for Technologists, Physics, Biology and Protection. 10th ed. St Louis, MO: Elsevier Mosby; 2013.

44. Sechopoulos I, Bliznakova K, Qin X, Fei B, Feng SS. Characterization of the homogenous tissue mixture approximation in breast imaging dosimetry. Med Phys. 2012;39(8):5050-5059. doi:10.1118/1.4737025.

45. Feng SS, Sechopoulos I. Clinical digital breast tomosyn-thesis system: dosimetric characterization. Radiology. 2012;263(1):35-42. doi:10.1148/radiol.11111789.

46. Sechopoulos I, Sabol JM, Berglund J, et al. Radiation dosimetry in digital tomosynthesis: report of AAPM Tomosynthesis Subcommittee Task Group 223. Med Phys. 2014;41(9):091501. doi:10.1118/1.4892600.

47. Klein R, Aichinger H, Dierker J, et al. Determination of aver-age glandular dose with modern mammography units for two large groups of patients. Phys Med Biol. 1997;42(4):651-671.

48. Hammerstein GR, Miller DW, White DR, Masterson ME, Woodard HQ , Laughlin JS. Absorbed radiation dose in mammography. Radiology. 1997;130(2):485-491. doi:10.1148/130.2.485.

49. Barkhausen J, Achim R, Schäefer F. Digital Breast Tomo-synthesis. Stuttgart, Germany: Thieme Publishers; 2016.

25. Sechopoulos I. A review of tomosynthesis. Part 1. The image acquisition process. Med Phys. 2013;40(1):014301. doi:10.1118/1.4770279.

26. Glynn CG, Farria DM, Monsees BS, Salcman JT, Wiele KN, Hildebolt CF. Effect of transition to digital mammography on clinical outcomes. Radiology. 2011;260(3):664-670. doi:10.1148/radiol.11110159.

27. Parikh R, Mathai A, Parikh S, Chandra Sekhar G, Thomas R. Understanding and using sensitivity, specificity and pre-dictive values. Indian J Opthalmol. 2008;56(1):45-50.

28. Ong MS, Mandl KD. National expenditure for false-positive mammograms and breast cancer overdiagnoses estimated at $4 billion a year. Health Aff (Millwood). 2015;34(4):576-583. doi:10.1377/hlthaff.2014.1087.

29. Iverson C, Christiansen S, Flanagin A, et al. Glossary of pub-lishing terms. In: AMA Manual of Style: A Guide for Authors and Editors. 10th ed. New York City, NY: Oxford University Press; 2007:935-966.

30. Pisano ED, Hendrick RE, Yaffe MJ, et al. Diagnostic accura-cy of digital versus film-mammography: exploratory analysis of selected population subgroups in DMIST. Radiology. 2008;246(2):376-383. doi:10.1148/radiol.2461070200.

31. Nields MW. FDA & digital mammography: why has FDA required full field digital mammography systems to be regulated as potentially dangerous devices for more than 10 years? Acad Radiol. 2010;17(5):652-657. doi: 10.1016/j.acra .2010.01.012.

32. van Ravesteyn NT, Van Lier L, Schechter CB, et al. Transition from film to digital mammography: impact for breast cancer screening through the national breast and cervical cancer early detection program. Am J Prev Med. 2015;48(5):535-542. doi:10.1016/j.amepre.2014.11.010.

33. Computer-aided detection mammograms finds cancers ear-lier but increases risk of false positives. BreastCancer.org website. http://www.breastcancer.org/research-news/2013 0425-4. Published April 25, 2013. Accessed March 24, 2016.

34. Fenton JJ, Xing G, Elmore JG, et al. Short-term outcomes of screening mammography using computer-aided detection: a population-based study of Medicare enrollees. Ann Intern Med. 2013;158(8):580-587. doi:10.7326/0003-4819-158-8 -201304160-00002.

35. PACS: an ounce of prevention. U.S. Food & Drug Administration website. http://www.fda.gov/Radiation -EmittingProducts/MammographyQualityStandardsAct andProgram/FacilityScorecard/ucm474994.htm. Accessed February 28, 2016.

36. Under MQSA, what medical records are facilities required to maintain and for how long should these records be kept? U.S. Food & Drug Administration website. http://www.fda


CEDirected Reading

Johnson

in the screening environment. J Clin Imaging Sci. 2014;4:9. doi:10.4103/2156-7514.127838.

62. Bidlek M, Kovács E, Fehér K, Gõdény M. New trends and novel possibilities in the diagnostic imaging of the breast [in Hungarian]. Magy Onkol. 2015;59(1):44-45.

63. Niklason LT, Christian BT, Niklason LE, et al. Digital tomo-synthesis in breast imaging. Radiology. 1997;205(2):399-406. doi:10.1148/radiology.205.2.9356620.

64. Baldwin P. Digital breast tomosynthesis. Radiol Technol. 2009;81(1):57M-74M.

65. Frequently asked questions on digital mammography. U.S. Food & Drug Administration website. http://www.fda.gov /radiation-emittingproducts/mammographyqualitystan dardsactandprogram/consumerinformation/ucm113973 .htm. Revised June 2, 2015. Accessed January 10, 2016.

66. Hologic receives FDA approval for first 3-D digital mammog-raphy (breast tomosynthesis) system [news release]. Bedford, MA: Hologic; February 11, 2011. http://investors.hologic .com/index.php?s=43&item=345. Accessed January 16, 2016.

67. Hu YH, Zhao W. The effect of amorphous selenium detec-tor thickness on dual-energy digital breast imaging. Med Phys. 2014;41(11):111904. doi:10.1118/1.4897244.

68. Alakhras M, Bourne R, Rickard M, Ng KH, Pietrzyk M, Brennan PC. Digital tomosynthesis: a new future for breast imaging? Clin Radiol. 2013;68(5)e225-e236. doi:10.1016/j .crad.2013.01.007.

69. Greer LR. The benefits of using synthesized 2D (C-view) images in breast tomosynthesis exams. Applied Radiology website. http://www.appliedradiology.com/articles/the -benefits-of-using-synthesized-2d-c-view-images-in-breast -tomoynthesis-exams. Published November 5, 2014. Accessed March 2, 2016.

70. Vecchio S, Albanese A, Vignoli P, Taibi A. A novel approach to digital breast tomosynthesis for simultaneous acquisition of 2D and 3D images. Eur Radiol. 2011;21(6):1207-1213. doi:10.1007/s00330-010-2041-y.

71. Zuley ML, Guo B, Catullo VJ, et al. Comparison of two-dimensional synthesized mammograms versus original digi-tal mammograms alone and in combination with tomosyn-thesis images. Radiology. 2014;271(3):664-671. doi:10.1148 /radiol.13131530.

72. Hakim CM, Catullo VJ, Chough DM, et al. Effect of the availability of prior full-field digital mammography and digital breast tomosynthesis images on the interpretation of mammograms. Radiology. 2015;276(1):65-72. doi:10.1148 /radiol.15142009.

73. Skaane P, Bandos Al, Eben EB, et al. Two-view digital breast tomosynthesis screening with synthetically reconstructed projection images: comparison with digital

50. Olgar T, Kahn T, Gosch D. Average glandular dose in digital mammography and breast tomosynthesis. Rofo. 2012;184(10):911-918. doi:10.1055/s-0032-1312877.

51. Baptista M, Di Maria S, Figueira C, Orvalho L, Vaz P. Deter-mination of backscatter factors in breast tomosynthesis using MCNPX simulations and measurements. Radiat Prot Dosimetry. 2015;165(1-4):325-330. doi:10.1093/rpd/ncv102.

52. Sechopoulos I, Suryanarayanan S, Vedantham S, D’Orsi C, Karellas A. Computation of the glandular dose in digital tomosynthesis of the breast. Med Phys. 2007;34(1):221-232. doi:10.1118/1.2400836.

53. Paulis LE, Lobbes MB, Lalji UC, et al. Radiation exposure of digital breast tomosynthesis using an antiscatter grid compared with full-field digital mammography. Invest Radiol. 2015;50(10):679-685. doi:10.1097/RLI.0000000000000168.

54. Dance DR, Young KC, van Engen RE. Estimation of mean glandular dose for breast tomosynthesis: factors for use with the UK, European and IAEA breast dosimetry protocols. Phys Med Biol. 2011;56(2):453-471. doi:10.1088/0031-9155 /56/2/011.

55. McCullagh JB, Baldelli P, Phelan N. Clinical dose perfor-mance of full field digital mammography in a breast screen-ing programme. Br J Radiol. 2011;84(1007):1027-1033. doi:10.1259/bjr/83821596.

56. Elmore JG, Jackson SL, Abraham L, et al. Variability in inter-pretive performance at screening mammography and radi-ologists’ characteristics associated with accuracy. Radiology. 2009;253(3):641-651. doi:10.1148/radiol.2533082308.

57. Vedantham S, Karellas A, Vijayaraghavan GR, Kopans DB. Digital breast tomosynthesis: state of the art. Radiology. 2015;277(3):663-684. doi:10.1148/radiol.2015141303.

58. Tosteson AN, Fryback DG, Hammond CS, et al. Consequences of false-positive screening mammograms. JAMA Intern Med. 2014;174(6):954-961. doi:10.1001/jama internmed.2014.981.

59. Gilbert FJ, Tucker L, Gillan MG, et al. The TOMMY trial: a comparison of the tomosynthesis with digital mammogra-phy in the UK NHS Breast Screening Programme--a multi-centre retrospective reading study comparing the diagnostic performance of digital breast tomosynthesis and digital mammography with digital mammography alone. Health Technol Assess. 2015;19(4)1-136. doi:10.3310/hta19040.

60. Gilbert FJ, Tucker L, Young KC. Digital breast tomosyn-thesis (DBT): a review of the evidence for use as a screening tool. Clin Radiol. 2016;71(2):141-150. doi:10.1016/j.crad .2015.11.008.

61. Destounis S, Arieno A, Morgan R. Initial experience with combination digital breast tomosynthesis plus full field digi-tal mammography or full field digital mammography alone


CEDirected Reading


85. Andersson I, Ikeda DM, Zackrisson S, et al. Breast tomo-synthesis and digital mammography: a comparison of breast cancer visibility and BIRADS classification in a population of cancers with subtle mammography findings. Eur Radiol. 2008;18(12):2817-2825. doi:10.1007/s00330-008-1076-9.

86. Gur D, Bandos AI, Rockette HE, et al. Localized detection and classification of abnormalities on FFDM and tomosyn-thesis examinations rated under an FROC paradigm. AJR Am J Roentgenol. 2011;196(3):737-741. doi:10.2214/AJR.10.4760.

87. Helvie MA. Digital mammography imaging: breast tomo-synthesis and advanced applications. Radiol Clin North Am. 2010;48(5):917-929. doi:10.1016/j.rcl.2010.06.009.

88. Haas BM, Kalvra V, Geisel J, Raghu M, Durand M, Philpotts LE. Comparison of tomosynthesis plus digital mammog-raphy and digital mammography alone for breast cancer screening. Radiology. 2013;269(3):694-700. doi:10.1148 /radiol.13130307.

89. Kopans D, Gavenonis S, Halpern E, Moore R. Calcifications in the breast and digital breast tomosynthesis. Breast J. 2011; 17(6):638-644. doi:10.1111/j.1524-4741.2011.01152.x.

90. Sprangler ML, Zuley ML, Sumkin JH, et al. Detection and classification of calcifications on digital breast tomosynthe-sis and 2D digital mammography: a comparison. AJR Am J Roentgenol. 2011;196(2):320-324. doi:10.2214/AJR.10.4656.

91. Samei E, Thompson J, Richard S, Bowsher J. A case for wide-angle breast tomosynthesis. Acad Radiol. 2015;22(7):860-869. doi:10.1016/j.acra.2015.02.015.

92. Schell MJ, Yankaskas BC, Ballard-Barbash R, et al. Evidence-based target recall rates for screening mammogra-phy. Radiology. 2007;243(3):681-689. doi:10.1148/radiol .2433060372.

93. Preibsch H, Siegmann-Luz KC. Digital breast tomosynthe-sis [in German]. Radiologe. 2015;55(1):59-67. doi:10.1007 /s00117-014-2753-0.

94. Sumkin JH, Ganott MA, Chough DM, et al. Recall rate reduc-tion with tomosynthesis during baseline screening examina-tions: an assessment from a prospective trial. Acad Radiol. 2015;22(12):1477-1482. doi:10.1016/j.acra.2015.08.015.

95. Svahn TM, Houssami N, Sechopoulos I, Mattsson S. Review of radiation dose estimates in digital breast tomo-synthesis relative to those in two-view full-field digital mam-mography. Breast. 2015;24(2):93-99. doi:10.1016/j.breast .2014.12.002.

96. Cavagnetto F, Taccini G, Rosasco R, Bampi R, Calabrese M, Tagliafico A. ‘In vivo’ average glandular dose evalua-tion: one-to-one comparison between digital breast tomo-synthesis and full-field digital mammography. Radiat Prot Dosimetry. 2013;157(1):53-61. doi:10.1093/rpd/nct120.

97. Wallis MG, Moa E, Zanca F, Leifland K, Danielsson M. Two-view and single-view tomosynthesis versus full-field

breast tomosynthesis with full-field digital mammography images. Radiology. 2014;271(3):655-663. doi:10.1148/radiol.13131391.

74. FDA approves stand-alone 3D screening with Siemens tomo-synthesis platform [news release]. Siemens; May 18, 2016. https://usa.healthcare.siemens.com/press/pressreleases/heal thcare-news-2016-05-18-1. Accessed November 23, 2016.

75. Yee KM, Synthesized 2D images dramatically reduce DBT radiation dose. AuntMinnie website. http://www.auntminnie .com/index.aspx?sec=ser&sub=def&pag=dis&itemID=107 698. Published June 13, 2013. Accessed March 10, 2016.

76. Digital breast tomosynthesis (DBT) system. U.S. Food & Drug Administration website. http://www.fda.gov/Radiation -EmittingProducts/MammographyQualityStandardsActand Program/FacilityCertificationandInspection/ucm413117 .htm. Updated April 4, 2016. Accessed September 21, 2016.

77. Skaane P, Bandos AI, Gullien R, et al. Comparison of digi-tal mammography alone and digital mammography plus tomosynthesis in a population-based screening program. Radiology. 2013;267(1):47-56. doi:10.1148/radiol.12121373.

78. Ciatto S, Houssami N, Bernardi D, et al. Integration of 3D digital mammography with tomosynthesis for population breast-cancer screening (STORM): a prospective compari-son study. Lancet Oncol. 2013;14(7):583-589. doi:10.1016 /S1470-2045(13)70134-7.

79. Houssami N, Skaane P. Overview of the evidence on digital breast tomosynthesis in breast cancer detection. Breast. 2013;22(2):101-108. doi:10.1016/j.breast.2013.01.017.

80. Peppard HR, Nicholson BE, Rochman CM, Merchant JK, Mayo RC 3rd, Harvey JA. Digital breast tomosynthesis in the diagnostic setting: indications and clinical applications. Radiographics. 2015;35(4):975-990. doi:10.1148/rg.2015 140204.

81. Rose SL, Tidwell AL, Bujnoch LJ, Kushwaha AC, Nordmann AS, Sexton R Jr. Implementation of breast tomosynthesis in a routine screening practice: an observa-tional study. AJR Am J Roentgenol. 2013;200(6):1401-1408. doi:10.2214/AJR.12.9672.

82. Friedewald SM, Rafferty EA, Rose SL, et al. Breast can-cer screening using tomosynthesis in combination with digital mammography. JAMA. 2014;311(24):2499-2507. doi:10.1001/jama.2014.6095.

83. Yang TL, Liang HL, Chou CP, Huang JS, Pan HB. The adjunctive digital breast tomosynthesis in diagnosis of breast cancer. Biomed Res Int. 2013;2013:597523. doi:10.1155/2013/597253.

84. Poplack SP, Tostevan TD, Kogel CA, Nagy HM. Digital breast tomosynthesis: initial experience in 98 women with abnormal digital screening mammography. AJR Am J Roentgenol. 2007;189(3)616-623. doi:10.2214/AJR.07.2231.


CEDirected Reading

Johnson

digital mammography: high resolution x-ray imaging observ-er study. Radiology. 2012;262(3):788-796. doi:10.1148/radiol .11103514.

98. Waldherr C, Cerny P, Altermatt HJ, et al. Value of one-view breast tomosynthesis versus two-view mammography in diagnostic workup of women with clinical signs and symp-toms and in women recalled from screening. AJR Am J Roentgenol. 2013;200(1):226-231. doi:10.2214/AJR.11.8202.

99. Gur D, Zuley ML, Anello MI, et al. Dose reduction in digital breast tomosynthesis (DBT) screening using synthetically reconstructed projection images: an observer performance study. Acad Radiol. 2012;19(2):166-171. doi:10.1016/j.acra .2011.10.003.

100. Wittenberg T, Wagner F, Gryanik A. Towards a computer assisted diagnosis system for digital breast tomosynthesis. Biomed Tech. 2012;57(supp 1):223-226.

101. Sahiner B, Chan HP, Hadjiiski LM, et al. Computer-aided detection of clustered microcalcifications in digital breast tomosynthesis: a 3D approach. Med Phys. 2012;39(1):28-39. doi:10.1118/1.3662072.

102. Morra L, Sacchetto D, Durando M, et al. Breast cancer: computer-aided detection with digital breast tomosynthesis. Radiology. 2015;277(1):56-63. doi:10.1148/radi ol.2015141959.

103. Good WF, Abrams GS, Catullo VJ, et al. Digital breast tomosynthesis: a pilot observer study. AJR Am J Roentgenol. 2008;190(4):865-869. doi:10.2214/AJR.07.2841.

104. Dang PA, Freer PE, Humphrey KL, Halpern EF, Rafferty EA. Addition of tomosynthesis to conventional digital mam-mography: effect on image interpretation time of screening examinations. Radiology. 2014;270(1):49-56. doi:10.1148 /radiol.13130765.

105. Frequently asked questions about DBT and MQSA training requirements. U.S. Food & Drug Administration website. http://www.fda.gov/Radiation-EmittingProducts /MammographyQualityStandardsActandProgram/Facility CertificationandInspection/ucm447869.htm. Revised May 22, 2015. Accessed February 28, 2016.

106. Zuley ML, Bandos AI, Abrams GS, et al. Time to diagnosis and performance levels during repeat interpretations of digital breast tomosynthesis: preliminary observations. Acad Radiol. 2010;17(4):450-455. doi:10.1016/j.acra.2009.11.011.


Directed Reading Quiz

continued on next page

Digital Breast Tomosynthesis and Full-Field Digital Mammography

1. The American Cancer Society’s 2015 breast screening recommendations include all of the following guidelines except: a. annual clinical breast examinations by a

physician or qualified health care professional.b. annual screening for women begins at age 40

years.c. annual screening performed on women between

ages 45 and 54 years.d. biennial screening performed on women 55

years and older.

2. In recent years, there has been more clinical interest in women who have dense breast tissue because high breast density:a. is associated with an increased risk of death

among cancer patients.b. increases with age. c. is linked to breast cancer risk.d. increases the positive predictive value of a breast

lesion.

3. The results of the Digital Mammography Imaging Screening Trial in 2008 reported that full-field digital mammography (FFDM):a. was not as effective as film-screen

mammography.b. proved more effective in breast cancer detection

in women older than 65 years.c. improved cancer detection in women 49 years

and younger. d. improved recall rates in women 65 years and

older with fatty breasts.

4. Which of the following statements is true regarding computer-aided detection (CAD) of breast images?a. Adoption of CAD technology remains low.b. CAD uses software algorithms to recognize and

highlight possible abnormalities.c. It is difficult to add CAD to FFDM.d. Studies have shown that CAD decreases

diagnosis of ductal carcinoma in situ.

Read the preceding Directed Reading and choose the answer that is most correct based on the article.

To earn continuing education credit: Take this Directed Reading quiz online at asrt.org/drquiz. Or, transfer your responses to the answer sheet on Page 326M and mail to ASRT, PO Box 51870,

Albuquerque, NM 87181-1870.

* Your answer sheet for this Directed Reading must be received in the ASRT office on or before this date.

QUIZ ID: 17801-03

1.75 Category A creditsExpires February 29, 2020*

http://www.asrt.org/drquiz



continued on next page

9. The results of the Oslo Tomosynthesis Screening Trial published in 2013 demonstrated the use of FFDM-DBT compared with FFDM alone increased cancer detection and decreased: a. recall rates.b. breast dose.c. false-positive rates.d. artifacts.

10. DBT likely is superior to FFDM for:1. clearly portraying benign findings.2. reducing image artifacts.3. preoperative measurement of breast lesions.

a. 1 and 2b. 1 and 3c. 2 and 3d. 1, 2, and 3

11. In a 2014 study from Friedewald et al, investigators compared FFDM-DBT with FFDM alone. They noted a 20% increase in invasive cancer detection for women who had DBT-FFDM compared with FFDM alone.a. trueb. false

12. Which of the following factors affects the amount of breast dose received during a DBT study?

1. scanning parameters2. breast composition3. breast size

a. 1 and 2b. 1 and 3c. 2 and 3d. 1, 2, and 3

5. According to the Mammography Quality Standards Act, the mean glandular dose received from a mammography study acquired with either analog or digital imaging should not exceed ______ mGy per exposure.a. 0.2b. 0.3 c. 2.0 d. 3.0

6. According to the article, the 2 primary reasons for recalls following FFDM are:a. breast density and anatomical noise.b. superimposition of structures and artifacts.c. motion and lack of adequate compression.d. tissue overlap and lack of radiologist expertise.

7. Which of the following factors increases the likelihood of false-positive results in mammography?

1. patient’s first mammogram2. shorter intervals between screenings 3. lack of images from previous screening

examinations

a. 1 and 2b. 1 and 3 c. 2 and 3 d. 1, 2, and 3

8. When using digital breast tomosynthesis (DBT) for breast screening, FFDM images still are necessary partly because: a. some radiologists have little to no experience

with DBT interpretation. b. only 2-D images can load on most interpretation

display monitors.c. benign lesions are easier to detect with FFDM

images. d. radiologists must have them for comparison of

prior 2-D images.



13. One way to decrease the amount of breast dose received during DBT imaging is to use:a. less compression.b. synthesized images from the DBT data set.c. the maximum tomographic angle.d. less filtration.

14. On average, interpretation of FFDM-DBT imaging requires ______% more time than interpretation of FFDM alone.a. 12b. 24c. 47d. 53

Documents

Full-Field Digital Mammography and Digital Breast ... RADIOOGIC TECHNOOGY, January/February 2017, Volume 88, Number 3 CE Directed Reading Full-Field Digital Mammography and Digital