Practical Reviews

Radiology - February 28, 2015 - Vol. 42 - No. 9

Automated Ultrasound Texture Analysis Hints at the Future of Breast Ultrasound

Author(s): Ardakani AA, Gharbali A, Mohammadi A

Author Email: gharbali@yahoo.com

This article counts as 0.25 credits in Breast Ultrasound.

The addition of automated breast ultrasound provides a specificity of 100% in discriminating benign from malignant masses.

Background: Morphologic analysis of solid breast masses seen on ultrasound has been the basis for distinguishing between benign and malignant masses over the past 20 years. By analyzing the shape and margin of the mass, radiologists are able to gauge the likelihood of malignancy and to recommend biopsy with a great deal of accuracy. Generally speaking, the ultrasound texture of the lesion may be described; however, its influence on the potential for malignancy is not well understood. Texture can be defined as a combination of gray-level value, color, brightness, and homogeneity or heterogeneity of those features within the mass.

Objective: To use statistical analysis to determine features of ultrasound texture of breast masses to differentiate benign from malignant ones.

Methods: 32 solid breast masses comprised the study set. All masses had undergone tissue sampling, and pathology indicated 20 benign masses and 12 malignant masses. All masses had been imaged using the same ultrasound machine and transducer. Regions of interest were selected, and quantitative texture analysis was performed using proprietary software (MaZda version 4.6; Lodz University of Technology, Lodz, Poland). Ten texture features were analyzed to determine characteristics that provided the greatest discrimination between benign and malignant diagnoses.

Results: When quantitative texture analysis was employed to discriminate between benign and malignant masses, the sensitivity was 94.3% and the specificity was 100.0%. The overall accuracy was 98.7%.

Conclusions: Automated texture analysis using proprietary software may contribute to the ability of the interpreting radiologist to improve sensitivity and specificity when evaluating solid breast masses.

Reviewer's Comments: The findings suggest a novel new technique in establishing malignant potential of a solid breast mass seen on ultrasound. Perhaps we may see this automated technique with newer generation ultrasound machines in the near future. A major limitation of this study is the limited sample size, since only 32 masses were evaluated. More comprehensive data and reproducibility of the performance will have to be established before clinical implementation can be considered.(Reviewer–Basil Hubbi, MD).

Article Reviewed: Classification of Breast Tumors Using Sonographic Texture Analysis. Ardakani AA, Gharbali A, Mohammadi A: J Ultrasound Med; 2015;34 (February): 225-231.


There is no credit for this activity or the credit has expired.
Comments:

Be the first to leave a comment.
Please Login to post comments.

Stephen R. Baker, MD
Coordinating Editor

Program Director, Professor, and Chairman of Radiology
Associate Dean for Graduate Medical Education
New Jersey Medical School
Scotch Plains, NJ
Rutgers University – New Jersey Medical School
Department of Radiology
Newark, NJ
Reports no commercial interest

Humaira Chaudhry, MD
Assistant Professor
Rutgers University – New Jersey Medical School
Department of Radiology
Newark, NJ
Reports no commercial interest

Basil Hubbi, MD
Assistant Professor of Radiology and Director of Breast Imaging
Rutgers University – New Jersey Medical School
Newark, NJ
Reports no commercial interest

Vineet R. Jain, MD
Associate Professor of Radiology
Department of Radiology
Montefiore Medical Center
Bronx, NY
Reports no commercial interest


Abhishek Kumar, MD
Interventional Radiology Fellow
Mount Sinai Beth Israel Medical Center
New York, NY
Reports no commercial interest

Otha W. Linton, MSJ
Executive Director
International Society of Radiology
Potomac, MD
Reports no commercial interest

John C. Sabatino, MD
Assistant Professor
Rutgers University – New Jersey Medical School
Department of Radiology
Newark, NJ
Reports no commercial interest

Sebastian Sadowski, MD
Radiology Imaging Specialists
Evergreen Park, IL
Reports no commercial interest

Uma Thakur, MD, MSK
Attending Radiologist
Radiology Imaging Associates
Waldorf, MD
Reports no commercial interest

Method of Participation: To receive credit for this activity, you should listen to the audio, read the content, and take the online post-activity quiz. To complete the quiz for credit: If you have not previously registered at the site, click on “New Customer Registration” and follow the directions. You will need your account number (located above your name on the Table of Contents) and your mailing zip code. Once you’ve logged in, click on the “Take a Quiz” link. Enter the quiz code and select your answers. Once you click Submit, you will receive immediate notification of your score. Questions should be answered based on information presented in the issue, and 70% of questions must be answered correctly to pass the quiz and receive credit for the issue. To obtain documentation of activity participation, you are required to submit an online activity evaluation.

Media: Internet access to mp3 audio, web-hosted content, and PDF available.

Hardware/Software Requirements: The web-based CE activity is best viewed using a current-generation browser such as Internet Explorer 6.0 or higher or Firefox 3.6 or higher. The presentation component is designed for 1024×768 screen resolution or higher. Cookies must be enabled to take the self-assessment, complete the evaluation, and print the CME certificate. You may play or download audio content. To listen online, JavaScript must be enabled, and Flash is required.

Audio CD: Standalone CD player or computer CD-ROM drive and compatible software CD player.

PDF: Computer CD-ROM drive and Adobe Reader, http://get.adobe.com/reader/(or compatible) software.

Disclosure: Oakstone Publishing, LLC, has assessed conflicts of interest with its faculty, authors, editors, and any individuals who were in a position to control the content of this CME activity. Any identified relevant conflicts of interest were resolved for fair balance and scientific objectivity of studies used in this activity. Oakstone’s planners, medical reviewers, and editorial staff disclose no relevant commercial interests. Medical Proofer Dr W. Murray Yarbrough discloses stock in GILD.

Intended Audience: Radiologists and others interested in imaging and diagnostic radiology.

Special Prerequisites for Participants: There are no prerequisites for participants.

As is commendable by the Accreditation Council for Continuing Medical Education criteria, CME is integrated into the process for improving professional practice with educational strategies implemented to remove, overcome, or address barriers to physician change. Physicians are unable to consistently integrate new information into their practices because of the ways new medical knowledge is structured and disseminated, and because of the time it takes to completely scan for new information. Practical Reviews selects and translates peer-reviewed literature for practical use, therefore diminishing the barrier of time.

Learning Objectives: At the conclusion of this activity, participants should be better able to:

  • Discuss controversies, advantages and disadvantages of various modalities, including conventional radiography, MRI and CT, ultrasound, special procedures and interventional techniques.
  • Interpret reviews of the latest basic science findings and the impact of these on clinical problems.
  • Contrast non-conventional and conventional principles and practices and identify the best approach.
  • Demonstrate a working familiarity with the most important information and perspectives presented, and generally should be able to apply in practice the information supplied.
  • Examine evidence of the safety and efficacy of such approaches.
Accreditation: Oakstone Publishing, LLC, is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.

The American Board of Radiology accepts this CME-certified activity toward the AMA PRA Category 1 Credits™ requirement as established in the maintenance of certification criteria.

Designation for Practical Reviews: Oakstone Publishing, LLC, designates this enduring material for a maximum of 5 AMA PRA Category 1 Credits™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

Estimated time to complete each issue: 5 hours

This material is also eligible for 5 credit hours in Category 2-B of the Continuing Medical Education Program of the American Osteopathic Association, provided it is completed in its entirety.

Designation for Practical Reviews On-Demand: (Per Review) Oakstone Publishing, LLC, designates this enduring material for a maximum of 0.25 AMA PRA Category 1 Credits™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

Estimated time to complete each review: 15 minutes

This material is also eligible for 0.25 credit hours in Category 2-B of the Continuing Medical Education Program of the American Osteopathic Association, provided it is completed in its entirety.

Disclaimer: The opinions and recommendations expressed by the faculty and other experts whose input is included in this program are their own. This enduring material is produced for educational purposes only. Use of the Oakstone Publishing, LLC, name implies review of educational format design and approach. Please review the complete prescribing information of specific drugs or combinations of drugs, including indications, contraindications, warnings, and adverse effects, before administering pharmacologic therapy to patients.
FINANCIAL DISCLOSURE
Oakstone Publishing, LLC, has assessed conflicts of interest with its faculty, authors, editors, and any individuals who were in a position to control the content of this CME activity. Any identified relevant conflicts of interest were resolved for fair balance and scientific objectivity of studies used in this activity. Oakstone’s planners, medical reviewers, and editorial staff disclose no relevant commercial interests. Medical Proofer Dr W. Murray Yarbrough discloses stock in GILD. The following faculty report no relevant financial interests: Otha W. Linton and Drs Stephen R. Baker, Humaira Chaudhry, Basil Hubbi, Vineet R. Jain, Abhishek Kumar, John C. Sabatino, Sebastian Sadowski, and Uma Thakur.