| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Breast Imaging |
1 From the Division of Breast Imaging, Duke University Medical Center, Erwin Rd, Box 3808, Durham, NC 27710. Received June 28, 2001; revision requested July 23; revision received October 2; accepted October 10. Address correspondence to J.A.B. (e-mail: baker013@mc.duke.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Static images from 152 breast US examinations performed at 86 institutions were evaluated for compliance with ACR guidelines for breast US hardware, technical factors, imaging protocol, and image annotation. Official interpretations submitted by the referring facilities were compared with static images submitted by the facility. Discrepancies were confirmed by two dedicated breast radiologists after repeat imaging, short-interval follow-up imaging, or biopsy.
RESULTS: A total of 60.5% of cases did not comply with at least one ACR guideline on breast US and included 9.2% of cases with inadequate equipment, 14.7% of cases with inappropriate focal zone placement, at least 14% of cases with static images in only one imaging plane, and 25% of cases with incomplete patient identifiers. Clinically relevant interpretation errors and interpretation discrepancies were confirmed in 23 (15.1%) of 152 cases.
CONCLUSION: The majority of breast US examinations did not comply with at least some of the standards for quality set forth by the ACR. Attention to these basic standards could substantially improve image quality.
© RSNA, 2002
Index terms: Breast, ACR reporting and data system • Breast, US, 00.1298, 00.92, 00.94 • Images, interpretation, 00.1298 • Quality assurance, 00.1298
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Although interest in quality breast imaging has largely focused on screen-film mammography, ultrasonography (US) in the breast also plays a critical role in the diagnostic evaluation of screening-detected or palpable masses. US with a high-frequency transducer is essential for accurate noninvasive diagnosis of breast cysts and has shown promise in the differentiation of benign from malignant solid masses (12,13). In addition, US has been advocated and shown to be potentially useful in the examination of young or pregnant symptomatic patients (14), guidance of percutaneous needle interventions (15,16), and even in screening of some asymptomatic patients (17,18). Like screen-film mammography, however, breast US can be technically challenging and requires state-of-the-art equipment, with appropriate technical settings to create an optimal image. Furthermore, breast US is highly user dependent, as normal tissue can simulate a breast lesion while some breast cancers may be subtle. Finally, considerable variability has previously been demonstrated in the interpretation of breast sonograms (19,20).
Although no federal laws similar to the Mammography Quality Standards Act govern the practice of breast US, professional guidelines for performing high-quality breast US have been established by the American College of Radiology (ACR) standards (21). These guidelines specify equipment, technical factors, image annotation, and personnel "which should generally produce high-quality radiological care." However, no studies to our knowledge have evaluated the quality of breast US performed at modern breast imaging facilities. For this study, we reviewed breast sonograms obtained in diverse breast imaging facilities in patients who subsequently sought care at our tertiary care medical center. The purpose of this study was to evaluate the technical quality and imaging appearance of breast sonograms based on the ACR standards and to assess differences between initial interpretation of images and reinterpretation at this facility.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
One board-certified radiologist (J.A.B.) with fellowship training in breast imaging and 4 years of experience in breast US retrospectively reviewed each sonogram. First, the reviewer was blinded to screen-film mammograms, any further imaging performed at our institution, written interpretations provided with the studies, and final histopathologic or clinical outcomes of each case. Sonograms were evaluated to determine if they met the criteria of technical and image quality outlined in the ACR standards (21). Second, all available sonograms were compared with written interpretations of the sonograms from the referring institution, as well as mammograms and histopathologic information. Discrepancies between the original interpretation and the subsequent reinterpretation and/or biopsy outcome were noted. One of three fellowship-trained breast imagers (including M.S.S.) with 5, 8, or 9 years experience with breast US confirmed subjective evaluations of the image quality and interpretation discrepancies with the original outside reading. Final subjective decisions were made with consensus agreement among the observers.
Technical Evaluation of Image Quality
Static sonograms of the breast were evaluated to determine whether they satisfied the ACR guidelines for hardware, technical settings, and labeling, as listed in Figure 1.
|
The focal zone depth was compared with the depth of the imaged lesion or breast tissue. Appropriate focal zone depth was defined as being within 1 cm superficial to the anterior mass margin or within 1 cm deep to the posterior mass margin. Focal zones placed outside this range were recorded as either 1–2 cm too superficial or deep or more than 2 cm too superficial or deep.
Subjective assessments of the image gray scale and field of view were recorded and confirmed by a second experienced fellowship-trained breast imager. Image gray scale was subjectively evaluated and determined to be (a) appropriate, (b) inappropriate—high gain, or (c) inappropriate—low gain. Because the ACR standards describe only desired gray scale relative to cysts, and most sonograms in this study did not depict a cyst, appropriate gray scale was defined as images that displayed fat lobules as varying shades of gray. Sonograms on which fat lobules ranged from predominantly light gray to white were classified as inappropriate—high gain and would likely cause inappropriate echoes within simple cysts. Sonograms on which the parenchyma ranged from predominantly dark gray to black were classified as inappropriate—low gain, which could potentially mask legitimate low-level echoes in cysts or disguise solid, markedly hypoechoic masses as simple anechoic cysts.
In addition to the guidelines specified in the ACR standards, the field of view used to examine a lesion was also evaluated. Appropriate field of view was defined as one that included the entire mass along with sufficient superficial tissue to assess the depth of the mass from the skin and sufficient posterior tissue to assess posterior transmission of the ultrasound beam. Restricted field of view was defined as an entire examination that included all documented static images on which lesion depth, mass margins, or posterior sound transmission were not depicted. Sonograms of large masses that were simply larger than the footprint of the transducer were not included in the restricted category. Field of view was defined as excessive if more than 4 cm of tissue posterior to the mass or targeted tissue was included in the image, which resulted in an unnecessarily small image of the target area.
Each study was examined to determine whether images in two orthogonal planes were recorded. Images were examined for the presence of at least one caliper measurement of all clinically relevant abnormalities.
Image Documentation
Static images were reviewed for the labeling of anatomic location where the image was obtained. "Right" or "left" breast annotation was recorded and subsequently compared with that of available screen-film mammograms and written interpretations to determine if the right or left labeling was correct. Labeling of a specific anatomic position by using diagrammatic pictogram, breast quadrant, or clock annotation was noted. Any modifiers of the transducer position that indicated the distance from the nipple were also noted.
Each static image was examined for the notation of orientation (ie, transverse or longitudinal or radial or antiradial) of the US transducer relative to the patient, which was either written on the image in text format or noted on a diagrammatic pictogram. Note was made whether transducer orientation was annotated on all static images, more than half of static images, less than half of static images, or not indicated on any of the static images.
ACR standards indicate specific patient-identifying information that should be permanently documented on each US image. Each image was examined for permanent documentation of the patient’s first and last name, date of birth or other identifying number, date of study, name and location of the imaging facility, and identification of the sonographer or sonologist.
Image Interpretation
The official interpretation of the sonogram obtained from the outside interpreting institution or practice was available for review for 71 of the 152 sonograms. In each case, the static sonograms and any available mammograms from the referring institution were examined with the final interpretation from the referring facility. Additional sonograms were obtained at this facility when needed to clarify the findings from the original sonograms. Discrepancies between the official outside interpretation and the reinterpretation after repeat scanning at this facility were confirmed by a second dedicated breast imaging radiologist (including M.S.S.) and verified with biopsy and histopathologic evidence in 11 cases, additional US imaging with a state-of-the-art 13-MHz linear-array transducer (Elegra; Siemens Medical Systems, Issaquah, Wash) in six cases, or additional US imaging and at least 12 months of follow-up imaging in six cases. A discrepancy in interpretation was documented only if it had the potential to be clinically relevant and was classified in one of six categories: 1, characteristically benign mass interpreted as suspicious; 2, suspicious mass interpreted as typically benign; 3, normal fibroglandular tissue interpreted as a solid suspicious mass; 4, solid mass present but not perceived; 5, complex or solid masses interpreted as simple cyst; and 6, lesion at US did not correlate with mammographic finding.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Technical Evaluation of Image Quality
All 152 US examinations were performed with linear transducers. The transducer frequency could be determined for 151 examinations and ranged from 5 to 13 MHz. In 137 (90.7%) examinations, a transducer frequency of at least 7.0 MHz, or a center frequency of at least 6.0 MHz for broadband transducers, was used. Ten (6.6%) examinations were performed with a transducer frequency of less than 7.0 MHz. Thirty-three (21.8%) studies were performed with either a 7.0- or 7.2-MHz transducer. A 7.5-MHz transducer was used in 30 (19.9%) examinations, while a transducer frequency of more than 7.5 MHz was used in 39 (25.8%) examinations. A broadband transducer with a center frequency of at least 6 MHz was used in 35 (23.2%) examinations and of less than 6 MHz, in four (2.6%) examinations. Therefore, in 14 (9.3%) of 151 examinations, a transducer frequency of less than 7.0 MHz (10 cases) or broadband center frequency of less than 6.0 MHz (four cases) was used.
The depth of the focal zone was indicated on 150 sonograms. The focal zone was placed appropriately within 1 cm of the depth of the targeted tissue on 128 (85.3%) sonograms. The focal zone was placed between 1 and 2 cm deep to the targeted tissue on 17 (11.3%) sonograms and more than 2 cm beyond the targeted tissue on five (3.3%). Therefore, the focal zone was inappropriately positioned on 22 (14.7%) sonograms on which its position could be determined (Figs 2, 3). The focal zone was not set superficial to the targeted tissue on any sonograms.
|
|
|
|
A mass or other abnormality was identified on 119 of 152 sonograms. At least one static image with caliper measurement was documented in 116 (97.5%) examinations. Three (2.5%) abnormalities were not measured in any projection, while 28 (23.5%) were measured in only one imaging plane.
Image Documentation
On 150 (98.7%) of 152 sonograms, the left or the right breast imaged was correctly labeled. On 151 (99.3%) sonograms, information regarding the specific anatomic location where the static image was obtained, either in the form of clock position, quadrant position, or as depicted on a pictogram, was provided. Some indication of the distance of the lesion from the nipple was provided on 20 (13.1%) sonograms.
The patient’s first and last name were accurately labeled on 149 (98.0%) of 152 sonograms. On one sonogram, only the patient’s first name was listed, while patients’ last names were misspelled on two other sonograms. On 18 (11.8%) sonograms, a patient-identifying number such as date of birth or medical record number was not listed. The name of the imaging institution was included on 128 (84.2%) sonograms. On 38 (25%) sonograms, at least one of the three required patient identifiers—full name, identifying number, and name of the facility—was not accurately labeled. Although permanent documentation of the location of the imaging facility and the sonographer’s or sonologist’s name, initials, or identifying symbol are advised in the ACR standards, this information was absent from virtually all sonograms. This absence was not included in the final determination of the sonograms that met the guidelines of the ACR standards.
Overall ACR Criteria
In this study of 152 breast sonograms, compliance with all ACR guidelines in the performance of breast US examinations (except labeling of imaging facility location and sonographer or sonologist identifier) was documented in 39.5% of cases. Ninety-two (60.5%) cases did not fully meet all ACR standards with regard to US hardware used, technical factors selected, imaging projections, lesion measurement, image documentation, or patient identification. An additional 22 (14.4%) cases did not meet subjective standards for image quality owing to unsuitable gray-scale gain setting or inappropriate field of view.
Image Interpretation
In 23 (15.1%) of the 152 cases, substantial disagreements between the original interpretation and the subsequent interpretation at this institution were identified. Only disagreements in interpretation with the potential to affect patient outcome were included. Discrepancies were confirmed with biopsy and histologic evaluation in 11 cases, consensus of two dedicated breast imagers by using repeat mammographic and US studies in six cases, and repeat imaging with at least 12 months follow-up (range, 12–15 months) in six cases. Although the official interpretation was available for review in 71 of the 152 cases, virtually all of the remaining cases showed either no abnormality, obvious simple cysts for which the patient received no additional treatment, or a suspicious mass for which the patient presented for definitive biopsy.
Of the 23 sonograms with discrepancies in interpretation, five (21.7%) were identified on which a characteristically benign mass was interpreted as suspicious for malignancy, and biopsy was recommended. Three of these masses were reinterpreted at this institution as typical benign-appearing intramammary lymph nodes (Fig 5) with pathognomonic "horseshoe" or "target sign" appearance. Intramammary lymph node was not mentioned as a differential diagnosis in any of the official outside interpretations of these sonograms. Biopsy was recommended in a patient with silicone breast implants for a lesion with the characteristic appearance of extracapsular silicone (Fig 6). Core-needle biopsy performed at the request of the referring surgeon confirmed the diagnosis of foreign body giant cell reaction with abundant silicone. A final lesion in which imaging at the referring facility was performed in only the antiradial projection was interpreted as a suspicious solid mass. Repeat US at this institution demonstrated a dilated duct that appeared anechoic and elongated and extended to the nipple when imaged in the radial projection (Fig 7). Again, histologic examination after wire localization and surgical excision confirmed benign duct ectasia rather than a solid mass.
|
|
|
|
|
|
In nine (39.1%) of 23 cases, the referring facility reported a solid suspicious mass, which at evaluation of the static images was reinterpreted at this institution to likely represent normal fibroglandular tissue (Fig 8). When repeat high-frequency high-resolution US was performed at this institution, no mass could be identified. For eight of the nine cases, US examination at the referring institution was performed with transducer frequency of between 5 and 7.5 MHz. Field of view met the criteria for "excessive" in three of these nine cases, and in one case, imaging was performed in only one projection.
|
|
|
|
|
Each of the 23 cases with differing interpretations or histologically confirmed interpretation errors had the potential for substantial effect on the patient’s care. Unnecessary biopsies were recommended by the referring facility in 14 cases with either normal fibroglandular tissue (nine cases) or a typically benign mass (five cases), while two malignancies were interpreted as typically benign, and one malignancy was overlooked.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
However, the Mammography Quality Standards Act addresses only one modality of breast imaging: screen-film mammography. During the 1990s, breast US has steadily increased in importance in breast evaluation. Initially used only to help differentiate cysts from solid breast masses, breast US has acquired new roles due in part to substantial improvements in instrumentation. Higher frequency transducers and all-digital imaging systems have resulted in marked improvements in image resolution and contrast. On the basis of the fine detail of US images, authors of several studies (12,13) now advocate the use of US to help differentiate benign from malignant solid breast masses. US is widely used as a tool for guiding percutaneous interventions, which include cyst aspiration, core-needle biopsy, and wire localization. Some investigators (18) have demonstrated potential improvements in breast cancer detection by including US in the screening protocol in patients with mammographically dense parenchyma.
As increased use of mammography in the 1970s and 1980s prompted closer evaluation of the technical factors and image quality, increased importance and complexity of breast US warrant a similar examination of the practice quality. One measure of quality in breast US is the guidelines issued by the ACR in the form of ACR standards (21). The ACR standards address indications for performing breast US and appropriate qualifications and responsibilities of the personnel involved. ACR guidelines on US equipment, technical factors, imaging protocols, and required labeling of the static sonograms were evaluated in this study.
The ACR standards provide guidance on US equipment appropriate for performing high-quality breast examinations. The standards specify that for breast US, linear-array transducers should be used with at least 7-MHz frequency or with a center frequency of at least 6 MHz for broadband systems. Almost 10% of sonograms analyzed in this study failed to meet this minimum requirement. In some cases, transducer frequencies as low as 5 MHz were used. For those who use single-frequency transducers less than 7.0 MHz, we believe that low-frequency transducers substantially limit the fine image detail of the breast, which can result in errors of interpretation, and every effort should be made to upgrade such systems to the highest frequency transducer possible. In other cases in this study, a variable-frequency transducer was used and inadvertently set at the minimum available frequency rather than the maximum frequency available for the transducer, despite the location of the target in the near field. In cases in which the minimum frequency was used, the ACR standard was not met owing to either a lack of awareness of the need for maximum available frequency or inattentiveness to the technical details of the examination.
With the assumption that the required hardware is available, technical settings play a substantial role in accurate US imaging and interpretation. For example, the ACR standards stipulate that "The focal zone should be set at the depth of the lesion," because the focal zone provides the narrowest beam and the best possible resolution. Misplacement of the focal zone can be due to either a lack of understanding of its preferred placement or oversight of the technical factors of the study. Nevertheless, in almost 15% of cases in this study, the focal zone was placed more than 1 cm deep to either the posterior margin of the targeted lesion or, when no lesion was present, to the interface between retroglandular fat and pectoralis muscle. Appropriate placement of the focal zone may substantially improve image resolution at no additional cost.
The ACR standards provide additional guidance on technical factors that affect the breast sonogram. In accordance with the standards, "gain settings should be adjusted to allow simple cysts to be distinguished from solid masses ... [but] not so high that artifactual echoes are placed within simple cysts." Evaluation of gray-scale images is nevertheless subjective, which is a limitation of this study. Almost 5% of the outside breast sonograms in this study were judged to have gray-scale settings sufficiently hypoechoic or hyperechoic so that a lesion could be overlooked or misinterpreted. In these cases, normal fat lobules were inappropriately displayed as virtually anechoic or completely hyperechoic.
A protocol for high-quality breast US is outlined in the ACR standards and requires imaging and documentation of a lesion or the area of interest in at least two orthogonal planes. Imaging in orthogonal planes is particularly important in breast US, because breast lesions may be subtle and may be detected in only one projection and because normal breast anatomy, including fat lobules and lactiferous ducts, may mimic a lesion in only one plane. In fact, the ACR standards plainly stipulate that "one view is insufficient." In this study, imaging was performed in only one plane, or the image plane was not labeled in more than one of every six cases.
Although measurements of breast lesions in two orthogonal projections are not explicitly required, the ACR standards specify that "the maximal dimensions of a mass should be included." Almost all lesions in this study were measured in at least one imaging plane. However, in almost one-quarter of cases, the largest dimension in the second orthogonal plane was not documented. Routine measure of abnormalities in both planes increases the likelihood that the largest dimension will be recorded and increases the likelihood that imaging will be performed routinely in orthogonal projections. By routinely following a breast US protocol that includes imaging and measurements in two orthogonal views, a substantial number of breast sonograms could be improved or better documented at no additional cost to the patient and essentially at no additional cost in hardware or time to the imaging facility.
The most common deviation from the ACR standards in this study was a failure to adequately annotate the static images obtained from breast US. Although anatomic location was correctly documented in almost all examinations, appropriate identification of the patient and imaging facility on the sonograms was often inadequate. The ACR standards specify that the patient’s first and last name, identification number and/or date of birth, and facility name should be permanently recorded on the images. In 25% of cases in this study, one or more of these identifying labels were not listed. Each is important to prevent confusion regarding which patient underwent imaging and where the procedure was performed. An identifying number is particularly important in the event that two patients share similar names. Again, at no cost to the patient or imaging facility, inclusion of this information on every image can reduce the likelihood of error.
Although the primary goal of this study was to evaluate the technical quality of breast US as currently performed, official image interpretations were also reviewed. Substantial disagreements between the original outside interpretation and the subsequent reinterpretation at this facility were documented in at least 15.1% of cases. These cases included initially overlooked breast carcinoma, misdiagnosed typically benign masses, and normal breast parenchyma misclassified as suspicious masses.
Although the specific cases of interpretation discrepancy are disturbing, the 15.1% error rate is considered an estimate in this investigation for three reasons. First, cases in this study include only those patients who were referred or sought a second opinion at this institution. Therefore, the study population is more likely to include complicated cases or cases in which the patient was not confident of the reported result. Second, official interpretations from the facility at which the sonogram was obtained were available in only 71 of the 152 cases. Therefore, although clinically important discrepancies were found in 23 (32.4%) cases, many of the remaining sonograms showed obvious simple cysts, conspicuous suspicious masses, or no abnormality. To provide the most conservative estimate of the interpretation error rate, these sonograms were presumed to have been correctly interpreted, because the cases appeared straightforward and the patients received appropriate referral for clinical evaluation, routine screening, or biopsy. The resulting discrepancy rate is therefore estimated at 15.1% (23 of 152 cases). Third, only 11 of the 23 interpretation discrepancies were confirmed with biopsy. In the remaining cases, either immediate repeat imaging was performed for cases deemed pathognomonic or repeat imaging was performed with at least 1-year follow-up. Given these limitations, the calculated discrepancy rate of 15.1% is considered the minimum rate for the 152 sonograms included in this study. However, the complexity of these cases may not be representative of the complexity observed in routine clinical practice. A formal study of interpretation accuracy in breast US should be undertaken.
Nevertheless, interpretation discrepancies identified in this study raise concern regarding the quality of breast US interpretation performed in current clinical practice. Many of the cases suggest a lack of awareness of the expected appearance of benign structures, such as normal fat lobules, intramammary lymph nodes, and extracapsular silicone. Interpretation differences such as mistaking a dilated lactiferous duct for a solid mass suggest a lack of familiarity with optimal scanning technique, such as imaging in orthogonal planes, particularly in radial and antiradial projections where a lactiferous duct is most often evident. Less common, breast cancers were misinterpreted as benign findings or were not identified despite static images demonstrating the cancer. Some of the interpretation differences identified in this study may have been due to the referring radiologists’ evaluation of only static images rather than an actual observation or performance of the breast US. This effect has been studied previously in abdominal US (26) but to our knowledge has not been evaluated in breast imaging.
Overall, 60.5% of cases failed to meet at least some quality criteria of the ACR standards, and at least 15.1% of the cases had potentially important discrepancies in image interpretation. These findings suggest the need for renewed focus on the technical details of performing breast US. Interpretation discrepancies in this study—including two missed breast cancers—may signify the need for practitioners of breast US to update their interpretation skills. Clinically relevant discrepancies between the interpretation at the original imaging facility and reinterpretation at this facility also suggest the need to reduce interobserver or interinstitution variability. The voluntary ACR Breast Ultrasound Accreditation Program (www.acr.org/departments/stand_accred/accreditation/us_breast_bio_over .html) addresses some of these issues by specifying equipment requirements, personnel qualifications, quality control procedures, quality assurance activities, and evaluation of image quality. In an era of increased mandatory regulation in the field of breast imaging, radiologists engaged in breast US should carefully evaluate their present practices and consider the advantages of voluntary accreditation as a step toward addressing deficiencies.
|
FOOTNOTES |
|---|
Author contributions: Guarantor of integrity of entire study, J.A.B.; study concepts and design, J.A.B., M.S.S.; literature research, J.A.B.; clinical studies, J.A.B.; data acquisition, J.A.B.; data analysis/interpretation, J.A.B., M.S.S.; manuscript preparation, J.A.B., M.S.S.; manuscript definition of intellectual content, J.A.B.; manuscript editing, revision/review, and final version approval, J.A.B., M.S.S.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  E.-K. Kim, K. H. Ko, K. K. Oh, J. Y. Kwak, J. K. You, M. J. Kim, and B.-W. Park Clinical Application of the BI-RADS Final Assessment to Breast Sonography in Conjunction with Mammography Am. J. Roentgenol., May 1, 2008; 190(5): 1209 - 1215. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Kim, E.-K. Kim, J. Y. Kwak, B.-W. Park, S.-I. Kim, J. Sohn, and K. K. Oh Role of Sonography in the Detection of Contralateral Metachronous Breast Cancer in an Asian Population Am. J. Roentgenol., February 1, 2008; 190(2): 476 - 480. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Mesurolle, T. Helou, M. El-Khoury, M. Edwardes, E. J. Sutton, and E. Kao Tissue Harmonic Imaging, Frequency Compound Imaging, and Conventional Imaging: Use and Benefit in Breast Sonography J. Ultrasound Med., August 1, 2007; 26(8): 1041 - 1051. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. G. ODLE Breast Ultrasound Radiol. Technol., January 1, 2007; 78(3): 222M - 242M. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. A. Berg, J. D. Blume, J. B. Cormack, and E. B. Mendelson Operator Dependence of Physician-performed Whole-Breast US: Lesion Detection and Characterization. Radiology, November 1, 2006; 241(2): 355 - 365. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Costantini, P. Belli, R. Lombardi, G. Franceschini, A. Mule, and L. Bonomo Characterization of solid breast masses: use of the sonographic breast imaging reporting and data system lexicon. J. Ultrasound Med., May 1, 2006; 25(5): 649 - 659. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
