|Approvals/Requirements Satisfied by eRADIMAGING Courses|
|~ ASRT approval for ARRT Category A credit||~ MDCB approval by the Medical Dosimetrist Certification (Selected Courses)|
|~ ARMRIT accepted (All MRI Courses)||~ CAMRT and Sonography Canada recognize the ASRT approval (All Courses)|
|~ ARDMS accepted (All Courses)||~ Florida approval for all courses 1 credit or more|
|~ NMTCB accepted (All Courses)||~ California CE requirements met for all radiography courses|
|~ All Courses eligible of international radiographers' CPD requirements||~ All Courses available for RRAs|
|~ ASRT and MDCB are approved continuing education providers of ARRT and all courses are accepted by ARRT|
This sample course is for reference purposes only. It is not currently available for earning CE credits. To earn ARRT CE credits please subscribe to eRADIMAGING where you will see a complete listing of all active and eligible CE courses.
The Role of CT in Breast Imaging
Paige Pettigrew, BS, RT(R) and Jeff L. Berry, MS, RT(R)(CT)
*Radiographer, Oklahoma City, Oklahoma.
†Assistant Professor, Radiography Program Director, University of Oklahoma Health Sciences Center, College of Allied Health, Department of Medical Imaging and Radiation Sciences, Oklahoma City, Oklahoma.
Address correspondence to: Paige Pettigrew, BS, RT(R), Radiographer, 7902 Ponderosa Drive, Tuttle, OK 73089. E-mail: firstname.lastname@example.org.
Disclosure Statement: Ms Pettigrew and Mr Berry report having no significant financial or advisory relationships with corporate organizations related to this activity.
Recent studies on the use of computed tomography (CT) technology for breast imaging have refocused the radiology community's attention on the potential benefits of breast CT, after it was deemed not viable in the 1970s. Significant advances in technology have enabled breast CT to possibly become another alternative to mammography and magnetic resonance imaging for use in breast cancer screening. Currently, breast CT is being developed and researched with technology that is still in clinical trials and not commercially available; however, it has been recognized that CT may potentially play a major role in breast imaging by providing advanced treatment options for patients. This article discusses the components, features, advantages, disadvantages, and protocol of 2 breast CT systems—conventional whole-body scanner breast CT and the dedicated breast CT system—and demonstrates how CT may be a significant tool for breast imaging radiologic professionals. The cost and technological advances of CT make it a strong competitor to mammography in winning the battle against breast cancer.
Breast cancer is a pathology that causes 40 000 deaths each year in the United States, and is the second leading cause of death in women such that 1 out of every 8 women is afflicted with this life-threatening disease.1-3 Early detection is necessary to treat patients and improve their chance for survival. The number of deaths has been drastically reduced by 30% to 50% over the last 15 years due to technological advancements in mammography.1 Mammography is currently the gold standard for detecting breast cancer worldwide; however, there are complications.4,5 One major complication is the lack of sensitivity with mammography, which makes it impossible to detect some abnormalities, such as breast lesions and breast cancer, as indicated in a recently conducted multi-institutional trial funded by the American College of Radiology Imaging Network that showed 30% of cancers are not detected by mammography.6 In addition, studies illustrated 70% to 90% of biopsies conducted as a result of suspicious mammograms are negative, driving up medical costs and causing patients unnecessary stress.5-7 Other complications include not being able to visualize the breast in a 3-dimensional (3D) view and how the superimposition of breast tissues with mammography makes it harder to detect abnormalities.8,9
Recent advances in technology have shown that breast computed tomography (CT) systems are promising in early detection. Breast CT imagery provides a 3D image of the breast and reduces the superimposition10 of breast tissue, thus enabling improved tumor detection, using a fast data acquisition time, and providing inclusion of the entire volume of the breast.8 Early tumor detection has been shown to facilitate a more favorable prognosis for the patient.4 Additionally, the radiation dose in breast CT is equal to or less than the dose used in mammography; this differs from comparisons of other X-ray procedures to CT exposures.1 Breast CT may prove to be more accurate while also providing other advantages, such as being less intrusive than mammography and less intimidating than magnetic resonance imaging (MRI).
Experimental and research studies have centered on using breast CT as another diagnostic tool for breast cancer.11 Studies indicate breast CT can be as valuable as mammograms, if not more so. The radiation dose, sensitivity, specificity rates, how well spatial/contrast resolution are visualized, as well as the ability of breast CT to differentiate malignant lesions from benign lesions and the extent of the lesions are still being researched with promising results. This literature review focuses on the 2 kinds of breast CT imaging: breast CT with a conventional whole-body scanner and the dedicated breast CT system. The following important question arises: Will breast CTs become a more valuable imaging alternative to mammography and MRI for screening and treatment purposes?
The History to the Present
In 1975, the first dedicated breast CT system, the CT/M (mammography) scanner, was invented and consisted of a fan-beam geometry which produced 1-cm thick CT slices in 10 seconds.1 The woman laid prone on a table with an aperture for the breasts, which were imaged separately by submerging the breast into a container filled with running warm water. The results from studies concluded that this CT breast imaging had a high sensitivity in detecting malignancies; however, the specificity rate of 70% was not great, stemming from poor spatial resolution.1 Additionally, the high costs, the need for iodine contrast, and the high radiation dose of 120 kVp and 20 mAs1 compared to mammography were not acceptable. Consequently, the CT/M system was dismissed from the market.
Next, Muller et al researched breast CTs with a conventional whole-body scanner.12 Patients were imaged in the prone position with foam blocks for the breasts to suspend freely. The studies concluded that the accuracy was not very high, and that the CT/M dedicated breast CT produced far better results. In addition, with whole-body breast CT, the X rays passed through the whole thorax, which produced higher radiation doses. For instance, in a study by Miyake et al, the radiation dose was measured at 23.5 mGy, which was almost 10 times that of mammography.13 Thus, Muller et al deemed that whole-body breast CT should be limited to special cases because of this large radiation dose and high cost compared to mammography.14 There was one advantage to whole-body breast CT: when using this tool, it was possible to detect distant metastases through incidental findings. This occurred because whole-body breast CT images the entire thorax, improving the detection of these metastases compared to the dedicated breast CT system, which solely targets the breasts.
In the last decade, advancements were made in CT and computer technology. One example was the development of scanners capable of generating thinner slices that produced higher spatial resolution and contrast resolution than ever. The whole-body CT scanners, including multidetector CT (MDCT) scanners, for instance, were developed, and as with other newer imaging modalities, MDCT scanners produced faster acquisition time than single-detector systems.1,14 The MDCT scanner consisted of more detectors than a traditional CT scanner and they closely compressed together, which helps improve resolution and speed. The MDCT scanning process begins with the patient lying either prone or supine on the table. The machine scans the breasts from the level of the axilla to the inferior border of the breasts. One hundred milliliters of a nonionic iodinated contrast, iohexol, is injected prior to the scan to improve visibility of structures.14 It was noted in several studies that MDCT contrast-enhanced scanners may incorporate dynamic CT for distinguishing breast cancer from benign lesions. Inoue et al conducted a study using dynamic, contrast-enhanced MDCT to examine the characteristics of 173 breast lesions, of which 87% were malignant.1,15 The study results were that needlelike (spiculated) lesions contained a positive predictive value of 99% with MDCT, whereas mammography had a 70% to 80% positive predictive value.15 Therefore, the detection of tumors was much improved in these CT slices compared to mammography. Additional benefits of breast MDCT included rapid scanning speed, the ability to have the patient lay in a supine position which is ideal for surgical simulation, and the fact that both breasts were scanned at the same time, which permits the physician to make comparisons of the affected breast with the normal breast.14
Presently, with huge advancements in technology occurring constantly, both the whole-body conventional breast CT scanners and the dedicated breast CT systems are being tested, and to date, the dedicated breast CT system has produced superior results. The latest procedure for the dedicated breast CT system with a flat-panel detector begins with the patient lying prone on the table with an opening for one breast similar to the image shown in Figure 1.7 The breast hangs freely and is supported to reduce movement. The advanced gantry system may allow for excellent coverage of the breast tissue in close proximity to the chest wall. Each breast is imaged separately, and located under the table are an X-ray tube and a flat-panel detector that spins around the breast, producing cone-beam projection images.1 Five hundred cone-beam projection images are obtained 360° around the patient's breast in 16.6 seconds at 30 frames per second according to Lindfors et al.16 Eventually, a second system was constructed with a faster scan time of 9 seconds, which helped to reduce the incidence of motion artifact. The uncompressed breast CT slice images are similar to the image illustrated in Figure 2.4
Whole-Body Conventional Breast CT vs Dedicated Breast CT System
There are many advantages of the dedicated breast CT system over the whole-body conventional breast CT scanner, such as the reduced radiation dose and higher spatial resolution briefly mentioned in the previous section. The higher spatial resolution improves the visualization of microcalcifications,1 whereas with previous CT whole-body scanners, microcalcifications were not clearly visualized. However, currently microcalcifications are better visualized using mammograms.1 Glick et al stated many cases of nonpalpable carcinomas are detected only by microcalcifications, making microcalcifications very significant because they aid in determining malignancy.1 Tumor detection with dedicated breast CT is better visualized as well due to the improved spatial resolution in CT slices from reconstruction and the ability to decrease the background clutter that is normally seen in mammograms.1 Another benefit is that because less tissue is penetrated by X rays, a lower kVp setting is used, which means the breast tumor contrast will be higher compared to conventional breast CT.1 The signal-to-noise ratio will then be higher as a result of the flat-panel dedicated breast CT system producing lower kVp and higher tumor contrast.1 Although, the radiation dose is high compared to mammography and dedicated breast CT systems, whole-body conventional scanners do have some advantages, such as the whole thorax is able to be imaged, which enables easier detection of distant and local metastases. Whole-body conventional scanners are also found in most radiology departments and therefore are more widely available.1
Some disadvantages of dedicated breast CT systems include the inability to display widespread and local metastases equal to whole-body conventional scanners and that they are not widely available yet. Additionally, there are several design issues surrounding dedicated breast CT systems.1 One issue is the ability to image the chest wall. Sometimes ductal and glandular breast tissue may extend into the chest wall and axilla; with mammography, compression pulls the breast away from the chest wall enough to allow for visualization of the breast tissues.1 To address this issue, Boone et al employed a flexible neoprene hammock that stretches over the table and contains an opening for the breast, enabling easier visualization of the breast tissues.17,18 Compared to whole-body conventional breast CT scanners, the spatial resolution of dedicated breast CT systems is higher, which improves detection; however, mammography visualizes microcalcifications more clearly due to its higher spatial resolution, which reconstructed CT slices with a flat-panel detector cannot yet achieve.1 This could change as the development of dedicated breast CT systems progresses. Although there are some disadvantages, studies are currently being conducted and researched to investigate various issues in order to improve this new technology.
Detectors, X-ray Tubes, Image Reconstruction, and Volumetric Segmentation
Detectors used in the dedicated breast CT systems are usually indirect flat-panels that are composed of a cesium-iodine phosphor attached to thin film transistors and photoiodides on top of an amorphous silicon substrate.1 Alternative detector panels can be used, such as a low-noise detector which consists of multiple electron-multiplying charge-coupled devices that may produce additional benefits, including low electronic noise, wide dynamic range, negligible lag, and high frame rates.1 The detectors also allow for high spatial resolution and low image lag. Spatial resolution is high because of the detector material (indirect cesium-iodine phosphor), which aids in producing high quality images. Image lag is described as the carryover from one image to another when 2 exposures are taken very closely in time.1 Low image lag speeds up acquisition time and reduces artifacts.
The X-ray tube and image reconstruction algorithms used in the dedicated breast CT system are also very significant in producing optimal quality images. Glick et al stated that for dedicated breast CT systems, the X-ray tube should be small and strong enough to produce many projections in a small amount of time. Small X-ray tubes possess a high heat-load capacity1 which is excellent because it stores and absorbs heat without damaging the X-ray tube anode. Furthermore, CT reconstruction is accomplished by combining all the projections from various positions in order to construct a 3D view. Specific mathematical computations are required to obtain cone-beam reconstruction for breast CT systems so that 3D images can be produced that are of diagnostic quality. Cone-beam breast CT is essentially the same concept as a dedicated breast CT system; the use of the cone-beam means that only the breasts are exposed to X rays, decreasing radiation exposure to the body unlike the whole-body conventional breast CT. The most commonly used algorithm is filtered backprojection.1 Using advanced reconstruction techniques, a 3D view of the digital breast can be imaged in the axial, coronal, and sagittal planes.18,19
Another element to consider with breast CT is volume segmentation. This refers to how breast volume can be separated into different tissues by its density and spatial geometry10 similar to the illustration in Figure 3.4 This is successfully completed by using algorithms. When the breast tissue is segmented, the tumor and the tissue can be distinguished from each other and selected out of image slices that compose the volume set. A drawback to using volume segmentation is the low spatial resolution of the images and the increased radiation dose; however, Chen and Ning stated this will be improved as the CT scanner and algorithms advance. Due to reconstruction, volume segmentation, and other factors, they asserted that breast tumors will be detected more accurately by 3D breast imaging, and despite these drawbacks, breast tumor detection should involve breast CTs.10
One of the main concerns with regard to breast CT is that the radiation dose is higher than that of traditional mammography. In order to evaluate this concern, phantom simulation studies were conducted by Boone et al and Thacker et al in 2004 to calculate the mean glandular dose (MGD) of radiation in dedicated breast CT systems.21,22 The MGD is the average radiation dose to the breast glandular tissue. These studies showed that the radiation dose to the breast was more uniform than in mammography, probably due to breast CT having a more consistent coverage throughout the breast.21,22 The Monte Carlo phantom studies used by Boone et al and Thacker et al verified the dose to organs outside the X-ray field is slightly higher than in mammography due to higher X-ray energy and the use of different projection geometry. Even though the radiation dose is higher, it is still very low compared to other radiographic and CT procedures.23 However, the use of phantom studies to assess radiation dose measurement is to some extent questionable because it was not calculated using human tissue and therefore, the results may be hard to compare.
For approximately 20 years, the amount of radiation dose, for breast CT, was considered too high and ultimately discontinued. In 2001, Boone et al conducted a research using cadaveric breasts and breast phantoms on a dedicated breast CT system to determine the amount of radiation dose accumulated. Radiation dose was calculated by utilizing breast geometry with diameters ranging from 6 to 16 cm which are the usual breast sizes.24 Many mathematic formulas and published index doses were used to compute the mean radiation dose levels. Radiation dose was also calculated by estimating the mAs needed at each kVp to produce excellent quality images. Additionally, to test for dose homogeneity, the dose distribution was computed by 1 mm by 1 mm by 20 cm voxels.24 The study concluded breast CT involved the same amount of radiation dose or lower as mammography, and also found the doses were reduced compared to mammography for breasts that are thicker than 5 cm.24 There was an inconsistency reported by Lindfors et al, in which women with denser breasts were exposed to a greater radiation dose.16 This is primarily due to the fact that the breast CT scanner used in Lindfors' study incorporated technique factors that were increased to maintain adequate noise levels. Women with larger and thicker breasts are exposed to a higher radiation dose in mammography also as a result of increased technique factors.16
It is important to note that there are weaknesses in this model because the data are experimental, conducted on phantoms, and the accuracy of the mathematical formulas used are questioned because they incorporated numerous formulas.
The image quality from breast CT is excellent. Sometimes mammograms are hard to read due to background clutter from superimposing tissue,1 but the flat-panel detector geometry used with CT significantly reduces this background clutter. It produces improved spatial resolution and contrast resolution, allowing for better visualization of tumors. CT images parenchymal structures more clearly, thus enabling reduction of artifacts that are usually visualized in mammograms.1 Additionally, as noted by Glick et al, screening women with breast implants in mammography is challenging and time consuming1 but with CT breast imaging, it is much quicker. This is because in mammography, an implant is superimposed over the native breast tissue, making lesion detection harder to accomplish. Thus, with breast implants, breast CT provides a clearer image of lesions in this group of women. Mammography can be a more time-consuming method for imaging breast implants because the technologist has to use the Ekland maneuver to separate the breast tissue from the implant in order to achieve diagnostic images; breast CT produces a 3D view of the breast, eliminating this challenge. There are so many women who have breast implants, and utilizing breast CT will benefit these women.
The effectiveness of breast CT and the images it produced were analyzed in a study conducted by Boone et al at University of California Davis.10 Breast CT data were collected from 55 women using a dedicated breast CT system. Three hundred 512 by 512 images for each breast were produced. It was determined that breast CT had roughly the same radiation dose as mammography, showed fine anatomical detail, good microcalcification depiction, and excellent visualization of the tumor's soft tissue when contrasted against adipose tissues.10
According to an evaluation by Glick et al in 2007, the breast CT system requires a very low exposure per projection view.25 Due to the very low exposure, image noise can increase and affect image quality. Thus, there is a trade-off between patient dose and image noise in that higher doses will produce higher quality signals and reduced noise. A lower tube voltage will produce less X-ray photons and increased noise levels; however, microcalcification contrast will increase.1 Currently, this relationship between the tube voltage and the signal-to-noise ratio requires additional studies.
Scatter radiation may also affect image quality by creating artifacts in breast CT systems with flat-panel detectors that result in image degradation. In breast CT, the X-ray cone beam strikes a detector that is up to 40 cm in height. Scatter increases with the increasing height of the detector and may cause decreased contrast and artifacts in the images.1 Glick et al proposed using a grid to prevent scatter, but the radiation exposure is already low.1 Utilizing a grid can decrease X-ray photons and lower the radiation dose, ultimately producing unwanted higher noise. Another alternative suggested by Ning et al was to use a beam-stop array that estimates the scatter distribution on a sampled lattice.26 The beam-stop array technique is combined with a filter technique and requires 2 cone-beam projections at each projection angle to calculate the remaining scatter.26 One projection is used with a beam-stop array for estimating scatter and the other is used without the beam-stop array to acquire the scatter plus the initial image.26 The latest beam-stop array scatter correction algorithm consists of obtaining several projections, estimating the scatter produced on the projection images, and finally calculating the scatter corrected projection, which ultimately results in decreased scatter production.26 Even though this would reduce scatter, it necessitates a slight increase in exposure and examination time. As these studies show, there are always trade-offs (such as increasing the kVp and changing other parameters to compensate); therefore, more research is needed to study the effects and methods for prevention of scatter radiation.
Using cone-beam technology for breast CT has been shown to benefit image quality. A simulation study conducted by Chen and Ning showed the feasibility of cone-beam dedicated breast CT systems in identifying lesions only a few millimeters in size and calcifications 100 micrometers in size.9 The breast CT can produce better low contrast images which aide in detecting breast masses. A precise measurement of the location and volume visualization of lesions are possible as well as optimal resolution for calcification with high contrast. This is an important aspect of breast CT because physicians are now able to localize the lesions, which will lead to advanced treatment options with CT, such as using breast CT for biopsies.
In some cases, image quality is better for some breast lesions on whole-body breast CT than mammography, specifically if the breasts are dense or if the lesion is found on the chest wall or axilla.27 When using mammography, it is harder to see lesions on dense breasts due to the type of breast tissue and superimposition of structures. Breast tissue contains glandular tissue, connective tissue, skin, and fat as noted in Figure 4.4 During puberty, the breast is composed mainly of glandular tissue and after menopause, it changes to adipose.4 As women age, adipose makes it easier for mammography to interpret abnormalities. As stated by Nelson et al, in mammography, it is a general assumption that the breast is made up of 50% fat and 50% glandular tissue.4 This assumption of tissue classification provides dosimetry estimates and techniques to use for producing images. Nelson et al explained that tissue classification in mammography is unnecessary due to the superimposition of structures, the outline of a breast tumor can easily be obscured within the glandular tissues of the breast, and superimposed tumors can be overlooked even by the most skilled radiologist.4 In addition, sometimes mammography does not cover the entire chest wall or axilla, making CT a superior diagnostic tool for detecting lesions on the chest wall and axilla. Also, contrast may be used in breast CT to improve the detection of tumors. The experimental study by Glick et al showed breast lesions of 3 to 5 mm should be visible with high resolution using dedicated breast CT systems. This is not 100% proven because phantoms were utilized in that study. Further research needs to be conducted on human patients when breast CT becomes available for use.
In addition, a new advanced method for improving image quality for cone-beam breast CTs using a volume of interest (VOI) scanning technique was discussed by Chen et al.3 This includes placing a filtering mask between the X-ray source and the breast in image acquisition mode as illustrated in Figure 5.3 There is an aperture in the mask that permits exposure to the VOI and a lower filtered exposure to the area outside the VOI. The purpose of the VOI scanning technique is to reduce the radiation dose to the breast without affecting the quality of the image. It was reported that the radiation dose was lowered both inside and outside the VOI. Although noise was a bit increased outside the VOI region, scatter radiation was reduced drastically both inside and outside the VOI which improved the image quality. The study also illustrated that using the VOI mask technique increased the contrast-to-noise ratio by approximately 45%, which is excellent. The VOI scanning technique seems encouraging, indicating there may be other advanced techniques such as this one to improve and perfect breast CT systems.
Many studies with breast CTs are done without contrast agents, although both the whole-body conventional breast CTs and the dedicated breast CT systems may use iodine contrast. For dedicated breast CT systems, Lindfors et al noted that more disease was visualized without using contrast16; however, according to Glick et al and Boone et al, contrast agents such as iodine greatly improve the visibility of imaged tumors.1,10 In Boone's study, 100 mL of iodine contrast was injected at a rate of 4 mL per second, and the images were obtained 100 seconds after the injection. The tumors are well visualized by the iodine contrast, accumulating in the interstitial spaces and enhancing. Boone et al incorporated several images in his study, which clearly illustrated how using contrast was very beneficial.10 The risk of reaction to the contrast is very low. If a reaction does occur, it is usually very minor and results in hives, warmness, or redness.
Breast CT vs Mammography
Mammography is currently the primary tool for detecting breast cancer because it is widely available and cost effective; however, it poses several challenges. Boone et al stated that the biggest downfall in mammography is its decreased sensitivity in women with dense breasts.24 Mammography obscures tumors in dense breasts due to the superimposed glandular structures of the breast whereas breast CT eliminates superimposition by representing the whole volume of the breast as shown in Figure 6.20 Chen and Ning specified mammography does not detect small tumors due to the reduced intensity contrast and stated that cone-beam breast CT's biggest advantage is its ability to produce a digital version of the breast that maintain configurations, such as the spatial geometry and other features of the breast as illustrated in Figure 7.8,20
In addition, breast compression is necessary for clear visualization in mammography screening. Compression produces soft tissue contrast by flattening the breast to reduce tissue overlap and scatter, as well as immobilizing the breast to prevent motion. According to Glick et al, compression draws the breast away from the chest wall, but dedicated breast CT systems are able to image the breast close to the chest wall.1 Unlike mammography, breast compression is not necessary with both breast CT systems, and therefore, there is no pain or discomfort to the patient.10 With this knowledge, the patient may be more inclined to undergo screening if breast CT is provided as an option. Approximately 50% of women screened with mammography experience moderate or greater discomfort.28 According to The American Cancer Society, women ages 40 and older should undergo yearly mammograms; however, only 50% of these women follow this recommendation.29 This is possibly due, in part, to the associated anxiety that has been attributed to the mammography examination.
Digital mammography systems have been invented, improving the sensitivity; however, it is more expensive, not readily available, and contains increased scatter radiation.2 Also, even with digital mammography, the overlapped breast tissues may still obscure tumors and slow detection.16 A computer simulation study carried out by Gong et al compared the lesion detection accuracy using digital mammography and flat-panel CT breast imaging.5 The simulation study consisted of tumors that were realistic and breasts that had a structured background. The results were that the 5-mm lesions in breast phantoms were detected better with CT breast imaging than with digital mammography.5 In that study, both breast CT and digital mammography had the same average glandular dose of 4 mGy. These results certainly supported the use of breast CT imaging. However, even though the computer simulation studies are a beneficial tool for the assessment of imaging systems, hopefully in the future the study will be conducted on human patients to give breast CT systems more credibility.
In the study by Boone and Nelson et al, they explained that CT is superior to mammography in contrast resolution by 10 times, and that breast CT images provide high spatial resolution that enables both identification and classification of breast tissue.4,24 Improved soft tissue contrast is also noted in this study. One negative that was noted was that interpretation of images consists of a shorter time in mammography compared to breast CT; however, as breast CT becomes more available and more radiologists are becoming familiar with this new tool, the interpretation time will shorten.
Lindfors et al constructed studies of 10 volunteers that were healthy and 69 women with lesions using a dedicated breast CT system and comparing it to mammography. In the initial study, the results for the breast CT scans in 9 out of 10 healthy volunteers were excellent as the images revealed fine detail of anatomy and contrast.16 Motion was visualized on one patient's scan. There were no complications with breast positioning or the patient comfort. The results for the second study with 69 women illustrated that 58 of 65 lesions seen on mammography were detected on breast CT (89% accuracy).16 Of the 7 lesions not detected on breast CT, 3 were malignant, 2 were microcalcifications, 1 was a mass, and the other was a cyst.16 Two lesions were seen on mammograms, but were not seen on breast CT because the lesions turned out to be artifacts16 proving breast CT is useful and has possibilities of being more accurate.
The study reported the mean lesion conspicuity score to be 5.4, meaning breast CT is about equal to mammography in detecting lesions. A limitation for this study is only one mammographer observed and studied the images. Even though this study was an initial clinical experience, it certainly would have been more credible if there were more radiologists involved. Another study with more radiologists involved should be conducted to further prove the significance of the effectiveness of breast CT scanning.
Lindfors et al also reported that the 10 healthy women and 69 women with lesions were scanned stated that lying on the breast CT unit was a little uncomfortable on their necks and that the tabletop was firm; however, 66% of the women rated the breast CT system a 10 out of 10, preferring it strongly to mammography. However, masses were detected better on breast CT images compared to mammography, but microcalcifications were not visualized as well. This may be attributed to the early generation dedicated breast CT system and will most likely improve with higher resolution CT systems. Also, the study used 80 kVp for the scans, so if lower voltage is used, the microcalcifications will be more easily identified with higher contrast. Unfortunately, as stated previously, if lower voltage is used, image noise will increase. Future breast CT systems developments should include improvements to these 2 issues of signal-to-noise ratio and tube voltage.
Breast CT vs Breast MRI
Breast MRI is another excellent diagnostic tool because it has a higher sensitivity compared to mammography; however, there are some drawbacks, such as the cost and lengthy scanning time.10 Breast MRI is not widely available, and some patients cannot undergo MRIs if they become claustrophobic or have a metallic implant, such as a pacemaker.30 Boone et al predicted that breast CT would be more practical, less expensive by approximately 50%, and can be performed quicker.10 In addition, breast CT has several advantages over breast MRI when performing biopsies because it can use metallic needles and does not have spatial distortion.
Studies by Shimauchi et al and Lindfors et al discussed how breast CT may become another alternative to MRI for screening women who are high at risk for breast cancer and staging the extent of their disease.14,16 Lindfors et al affirmed that the cost of breast CT is unknown, but it will be less expensive than MRI and the potential interpretation time for CT may be shorter than that for MRI.
In the study by Shimauchi et al, they evaluated the accuracy of MDCT and MRI in detecting intraductal components of breast cancer. Sixty-nine patients already diagnosed with invasive carcinoma underwent both contrast-enhanced MDCT and MRI. MDCT confirmed a sensitivity, specificity, and accuracy of 61%, 88% and 71%, whereas MRI confirmed a sensitivity, specificity, and accuracy of 75%, 88% and 80%, respectively. Generally, breast cancer is better assessed with breast CTs compared to mammography and sonography15; however, this study indicated that MRI was more sensitive in detecting an intraductal cancer in the breast.14 There were some limitations of the study, which may have skewed the MDCT results. These included the difference in the size of the intraductal component, which is dependent on the morphologic structure of the breast, and the interpretation of the images.14 Also, the radiologists excluded masses as benign. Another factor to consider is the population. Only women with invasive carcinoma that contained an intraductal component were studied; this might have had an effect on the cases with the intraductal component because the study was only comprised of palpable, symptomatic breast cancer and not screen-detected cases.14 In addition, the population did not include studies of noninvasive carcinoma.14 Currently, breast MRI is slightly superior to breast CT; however, as breast CT systems become more advanced and commercially available, that may change.
Although breast CTs are producing promising results, with the dedicated breast system appearing superior to whole-body conventional breast CTs, there are areas for further in-depth research. For instance, more research regarding the effects of the signal-to-noise ratio and tube voltage and scatter radiation is necessary. Also, determining whether or not iodine contrast agents should be used needs to be investigated further. Another issue is that radiologists are not familiar with this technology yet and will need training on how to interpret breast CT images. Because breast CT is in the clinical trial stage and is not commercially available, many studies are conducted on phantoms or simulations, but more translational research is needed.
Even with these considerations, the current literature shows that the potential advantages of breast CT systems far outweigh the disadvantages. These advantages include the ability to image the entire volume of the breast, the high spatial resolution, the superior contrast resolution, the elimination of superimposed breast tissues and structures resulting in improved tumor detection, and the fine anatomical detail produced with breast CT. Although dedicated breast CT utilizes ionizing radiation, the radiation dose is equal to or less than the dose used for mammography. Also, unlike mammography, breast CT does not need to use breast compression, which is a major benefit for the comfort of the patient.
Although breast CT systems are still in development, the study of breast CT is moving in a positive direction, and it has the potential to gradually change the field of breast imaging. If breast CT becomes commercially available in the future, it will likely play an important role in screening and early diagnosis. Will breast CT replace mammography or MRI? Breast CT may have a vital role in evaluating women with dense breasts, implants, and small cancers in dense breasts. Translational research should help to provide evidence-based guidelines for clinical implementation of breast CT. Lowering patient dose, while increasing the resolution needed to detect angiogenesis at the ductal level, will help to increase survival rates among women.
1. Glick S. Breast CT. Annu Rev Biomed Eng. 2007;9:501-526.
2. Lai C, Shaw C, Geiser W, et al. Comparison of slot scanning digital mammography system with full-field digital mammography system. Med Phys. 2008;35:2339-2346.
3. Chen L, Shaw C, Altunbas M, et al. Feasibility of volume of interest (VOI) scanning technique in cone beam breast CT—a preliminary study. Med Phys. 2008;35:3482-3490.
4. Nelson TR, Cervina LI, Boone JM. Classification of breast computed tomography data. Med Phys. 2008;35:1078-1086.
5. Gong X, Glick SJ, Liu B, et al. A comparison simulation study comparing lesion detection accuracy with digital mammography, breast tomosynthesis, and cone-beam CT breast imaging. Med Phys. 2006;33:1041-1052.
6. Pisano ED, Gatsonis C, Hendrick E, et al. Diagnostic performance of digital versus film mammography for breast-cancer screening. New Engl J Med. 2005;353:1773-1783.
7. Kerlikowske K, Grady D, Rubin SM, et al. Efficacy of screening mammography. A meta-analysis. JAMA. 1995;273:149-154.
8. Chen Z, Ning R. Why should breast tumor detection go three dimensional? Phys Med Biol. 2003;48:2217-2228.
9. Chen B, Ning R. Cone-beam volume CT breast imaging: feasibility study. Med Phys. 2002;29:755-770.
10. Boone JM, Kwan AL, Yang K, et al. Computed tomography for imaging the breast. J Mammary Gland Biol Neoplasia. 2006;11:103-111.
11. Breast CT reaches clinical testing: may improve on mammography. UC Davis Health System Web site. Available at: http://www.ucdmc.ucdavis.edu/newsroom/releases/archives/cancer/2005/breast_ct5-2005.html. Accessed September 26, 2009.
12. Muller JWT, vanWaes PF, Koehler PR. Computed tomography of breast lesions: comparison with X-ray mammography. J Comp Assist Tomogr. 1983;7:650-654.
13. Miyake K, Hayakawa K, Nishino M, et al. Benign or malignant? Differentiating breast lesions with computed tomography attenuation values on dynamic computed tomography mammography. J Comp Assist Tomogr. 2005;29:772-779.
14. Shimauchi A, Yamada T, Sato A, et al. Comparison of MDCT and MRI for evaluating the intraductal component of breast cancer. AJR Am J Roentgenol. 2006;187:322-329.
15. Inoue M, Sano T, Watai R, et al. Dynamic multidetector CT of breast tumors: diagnostic features and comparison with conventional techniques. AJR Am J Roentgenol. 2003;181:679-686.
16. Lindfors KK, Boone JM, Nelson TR, et al. Dedicated breast CT: initial clinical experience. Radiology. 2008;246:725-733.
17. Boone JM. 2004. Breast CT: its prospect for breast cancer screening and diagnosis. In: Karellas A, ed. RSNA Categorical Courses in Diagnostic Physics: Advances in Breast Imaging—Physics, Technology, and Clinical Applications. Chicago, IL: RSNA; 2004:165-177.
18. Boone JM, Kwan ALC, Nelson TR, et al. Performance assessment of a pendant-geometry CT scanner for breast cancer detection. SPIE Med Imag. 2005;5745:319-323.
19. Grable RJ, Ponder SL, Gkanatsios NA, et al. Optical computed tomography for imaging the breast: first look. Proc Soc Photo Opt Instrum Eng. 2000;4082:40-45.
20. Metheany KG, Abbey CK, Packard N, et al. Characterizing anatomical variety in breast CT images. Med Phys. 2008;35:4685-4694.
21. Boone JM, Shah N, Nelson TR. A comprehensive analysis of DgN(CT) coefficients for pendant-geometry cone-beam breast computed tomography. Med Phys. 2004;31:226-235.
22. Thacker SC, Glick SJ. Normalized glandular dose (DgN) coefficients for flat-panel CT breast imaging. Phys Med Biol. 2004;49:5433-5444.
23. Sechopoulos I, Vedantham S, Karellas A, et al. Monte Carlo and phantom study of the radiation dose to the body from dedicated ct of the breast. Radiology. 2008;247:98-105.
24. Boone JM, Nelson TR, Lindfors KK, et al. Dedicated breast CT: radiation dose and image quality evaluation. Radiology. 2001;221:657-667.
25. Glick SJ, Thacker S, Gong X, Liu B. Evaluating the impact of x-ray spectral shape on image quality in flat-panel CT breast imaging. Med Phys. 2007;34:5-24.
26. Ning R, Tang X, Conover D. X-ray scatter correction algorithm for cone beam CT imaging. Med Phys. 2004;31:1195-1202.
18. Kim SM, Park JM. Computed tomography of the breast: abnormal findings with mammographic and sonographic correlation. J Comput Assist Tomogr. 2003;27:761-770.
28. Dullum JR, Lewis EC, Mayer JA. Rates and correlates of discomfort associated with mammography. Radiology. 2000;214:546-552.
29. Study shows mammography reminders work. American Cancer Society Web site. Available at: http://www.cancer.org/docroot/NWS/content/NWS_1_1x_Study_Shows_Mammography_Reminders_Work.asp. Accessed September 10, 2009.
30. Nishino M, Hayakawa K, Yamamoto A, et al. Multiple enhancing lesions detected on dynamic helical computed tomography-mammography. J Comput Assist Tomogr. 2003;27:771-778.
Please note, due to a change in the ARRT's post-test requirements, we can no longer show you the answers to the questions that you got wrong once you complete the test.
Like what you see? Click here to join eRADIMAGING today.
|What did you think of this article?
The Role of CT in Breast Imaging
|»||Comment From: gayl||» Posted on: 07/22/2010 17:34 PM|
|great article! easy read, interesting|
|»||Comment From: ksmtwnn||» Posted on: 07/26/2010 22:11 PM|
|very good and informative|
|»||Comment From: BC Jones||» Posted on: 03/06/2011 11:16 AM|
|Well written, easy to follow.Recommend|
|There are 14 total comments: View All Comments|