FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3200830032> ?p ?o ?g. }

• W3200830032 endingPage "211788" @default.
• W3200830032 startingPage "211788" @default.
• W3200830032 abstract "HomeRadiologyVol. 301, No. 3 PreviousNext Reviews and CommentaryFree AccessEditorialBreast Density and TomosynthesisMartin J. Yaffe Martin J. Yaffe Author AffiliationsFrom the Sunnybrook Research Institute, Department of Medical Biophysics, University of Toronto and Ontario Institute for Cancer Research, Sunnybrook Health Sciences Centre, 2075 Bayview Ave, Room S657, Toronto, ON, Canada M4N 3M5.Address correspondence to the author (e-mail: [email protected]).Martin J. Yaffe Published Online:Sep 14 2021https://doi.org/10.1148/radiol.2021211788MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In See also the article by Gastounioti et al in this issue.Dr Yaffe is a senior scientist at Sunnybrook Research Institute and professor of medical biophysics at the University of Toronto. He has over 300 peer-reviewed publications on breast cancer and is an honorary fellow of the Society of Breast Imaging and member of the Order of Canada, that country’s highest civilian honor.Download as PowerPointOpen in Image Viewer In this issue of Radiology, Gastounioti et al (1) report on the performance of an algorithm that they developed for estimating volumetric breast density (VBD) from digital breast tomosynthesis (DBT) images. Their technique showed consistently stronger association between density and the risk of breast cancer than current measurements from two-dimensional digital mammography (DM). Given the increasing use of DBT, this may provide opportunities for more accurate assessment of breast cancer risk, potentially useful in optimizing screening regimens or monitoring preventive interventions.Breast density refers to the proportion of fibroglandular tissue to fat in the breast and can be estimated from mammograms either subjectively or quantitatively. It is a strong independent risk factor for the development of breast cancer (2). In addition, the detection sensitivity of mammography is markedly decreased in women with very dense breasts (3).The most commonly used method of assessing density is subjective classification of the mammogram on the American College of Radiology Breast Imaging Reporting and Data System (BI-RADS) density scale (4), which extends from A (breasts are almost entirely fatty) to D (breasts are extremely dense, which lowers the sensitivity of mammography). Detection sensitivity has been observed to decrease from 95% to 65% as categories progress from A to D (5). Density can also be measured on a continuous scale, as the fraction of the projected area of the breast covered by dense tissue or by fractional dense volume. Gastounioti et al report on a retrospective case-control study of women who underwent combination DM/DBT examinations (132 with unilateral cancers and 528 controls matched on age at time of screening examination and ethnicity). They found that VBD, estimated from DBT using their algorithm, provided a stronger correlation with cancer in the contralateral breast than area or volumetric assessment computed from DM.What are the practical uses of density measurement? Three come readily to mind. First, a reliable estimation of breast cancer risk could serve as a marker to monitor interventions intended to reduce risk. This would be most valuable if density is a true surrogate for risk, increasing and decreasing as risk changes. The question of surrogacy is still not fully resolved, however.The other applications are for optimizing and individualizing regimens for cancer detection. A marker for risk could inform decisions on the ages for individuals to begin and discontinue screening and the optimal interval. Currently, in some programs, individuals at elevated risk are candidates for more expensive, less accessible screening modalities such as breast MRI because of its increased sensitivity, especially for aggressive cancers (6).Additionally, an estimation of the probability of cancer masking based on density could guide recommendations for alternative or supplemental screening with US or contrast-enhanced imaging, where the detrimental effect of density on sensitivity would be less. For these last two applications to be acceptable, the predictive power of the test used for stratification would have to be very high.Consider the cancer risk application. To be useful, a “low-risk” signal must have an extremely high negative predictive value; the chances that a woman would develop cancer after being identified as low risk (and, therefore, receiving a less aggressive screening regimen) would have to be almost zero. Density information alone would not provide that level of discrimination. Many women with low-density breasts will develop breast cancer. Kerlikowske et al (7) found that 48.4% of invasive cancers occurred in women with nondense breasts (two lowest BI-RADS categories), and 7.1% occurred in the lowest category. Risk-based screening regimens will probably be realized through the development of more discriminative biologic markers or marker panels that might include density in combination with other factors, such as a polygenic risk score (8), to define the truly lowest-risk individuals as well as those at average, medium, and higher risk.Using density-based measures to identify women for whom supplemental screening would improve cancer detection sensitivity is also appealing. Both breast US and contrast-enhanced DM are effective in the dense breast. But, using density alone, it would be necessary to perform supplementary screening on almost half the women without cancer to ensure inclusion in the supplemental screening cohort of half of the women whose cancers would have been missed by mammography (ie, would have interval cancers) (7,9). This is simply not practical. Again, additional criteria, perhaps from multifactorial assessment of cancer risk, would be needed to obtain a more definitively selected population.The density masking problem was identified in two-dimensional mammography and may have been at least partially overcome with DBT, which by separating information into image sections tends to “unmask” lesions. How much masking persists with DBT? Is it restricted to only the densest breasts? If DBT does indeed supplant DM, answering these questions would help reframe the significance of density masking and the need for supplemental screening.The algorithm introduced by Gastounioti et al combines image feature analysis and machine learning, has impressive performance, and represents a step forward. Their claims of clear superiority of DBT over two-dimensional mammography are perhaps overstated, however. Although some studies found improved sensitivity and/or specificity, published results are quite variable. While increased cancer detection rates have been shown with DBT, the additional cancers found tend to be the less aggressive subtypes (10). There are very little data on the reduction of interval cancers.Like the physics-based VBD algorithms that operate on two-dimensional DM, the formation of quasi–three-dimensional DBT images and the inference of density from them also requires assumptions and approximations, some which may be hidden under the “magician’s cloak” of a machine learning algorithm. It is reassuring that in previous work, Gastounioti et al have shown their algorithm to agree well with MRI density measures and between craniocaudal and mediolateral oblique views of the breast.The current work leaves some other key questions unanswered. The evaluation was only performed at one time point, demonstrating that density in one breast is strongly associated with the synchronous presence of cancer in the contralateral breast. Will a measurement made some time before cancer detection be predictive of that future cancer? The density algorithm was implemented on only one type of tomosynthesis system. Given the enormous variation of image acquisition parameters across current commercial systems, how generalizable are these results?Finally, it would have been useful to gain some insight into why the authors’ algorithm provides improved risk prediction. Both DBT and VBD density assessment from DM rely on the same physics, the attenuation of x-rays. Will textural analysis of the DBT images provide stronger predictions? It will be exciting to see if Gastounioti et al can answer these questions as they move forward with this work.Disclosures of Conflicts of Interest: M.J.Y. disclosed grant to institution from GE Healthcare; is a shareholder of Volpara Health Technologies; is the principal of Mammographic Physics Inc.References1. Gastounioti A, Pantalone L, Scott CG, et al. Fully automated volumetric breast density estimation from digital breast tomosynthesis. Radiology 2021.https://doi.org/10.1148/radiol.2021210190. Published online September 14, 2021. Link, Google Scholar2. McCormack VA, Burton A, dos-Santos-Silva I, et al. International Consortium on Mammographic Density: methodology and population diversity captured across 22 countries. Cancer Epidemiol 2016;40:141–151. Crossref, Medline, Google Scholar3. Boyd NF, Guo H, Martin LJ, et al. Mammographic density and the risk and detection of breast cancer. N Engl J Med 2007;356(3):227–236. Crossref, Medline, Google Scholar4. Sickles EA, D’Orsi CJ, Bassett LW, et al. ACR BI-RADS Mammography. In: ACR BI-RADS Atlas, Breast Imaging Reporting and Data System.Reston, Va:American College of Radiology,2013. Google Scholar5. Destounis S, Johnston L, Highnam R, Arieno A, Morgan R, Chan A. Using volumetric breast density to quantify the potential masking risk of mammographic density. AJR Am J Roentgenol 2017;208(1):222–227. Crossref, Medline, Google Scholar6. Saslow D, Boetes C, Burke W, et al. American Cancer Society guidelines for breast screening with MRI as an adjunct to mammography. CA Cancer J Clin 2007;57(2):75–89. Crossref, Medline, Google Scholar7. Kerlikowske K, Sprague BL, Tosteson ANA, et al. Strategies to identify women at high risk of advanced breast cancer during routine screening for discussion of supplemental imaging. JAMA Intern Med 2019;179(9):1230–1239. Crossref, Medline, Google Scholar8. Wolfson M, Gribble S, Pashayan N, et al. Potential of polygenic risk scores for improving population estimates of women’s breast cancer genetic risks. Genet Med 2021.https://doi.org/10.1038/s41436-021-01258-y. Crossref, Medline, Google Scholar9. Alonzo-Proulx O, Mainprize JG, Harvey JA, Yaffe MJ. Investigating the feasibility of stratified breast cancer screening using a masking risk predictor. Breast Cancer Res 2019;21(1):91. Crossref, Medline, Google Scholar10. Chong A, Weinstein SP, McDonald ES, Conant EF. Digital breast tomosynthesis: concepts and clinical practice. Radiology 2019;292(1):1–14. Link, Google ScholarArticle HistoryReceived: July 14 2021Revision requested: July 27 2021Revision received: July 30 2021Accepted: Aug 4 2021Published online: Sept 14 2021Published in print: Dec 2021 FiguresReferencesRelatedDetailsAccompanying This ArticleFully Automated Volumetric Breast Density Estimation from Digital Breast TomosynthesisSep 14 2021RadiologyRecommended Articles Breast Cancer Risk Prediction Using Deep LearningRadiology2021Volume: 301Issue: 3pp. 559-560Tomosynthesis Is Taking Small Steps to Become the Standard for Breast Cancer ScreeningRadiology2021Volume: 299Issue: 3pp. 568-570Digital Mammography and Breast Tomosynthesis Performance in Women with a Personal History of Breast Cancer, 2007–2016Radiology2021Volume: 300Issue: 2pp. 290-300Addressing Racial Inequities in Access to State-of-the-Art Breast ImagingRadiology2022Volume: 306Issue: 2The Impact of Acquisition Dose on Quantitative Breast Density Estimation with Digital Mammography: Results from ACRIN PA 4006Radiology2016Volume: 280Issue: 3pp. 693-700See More RSNA Education Exhibits Breast Density Included in the Modern Rules of Mammographic ScreeningDigital Posters2019Let’s Talk about Next-Generation Breast Cancer Screening Programs: How Should We Do? What Should We Use?Digital Posters2020Contrast-Enhanced Mammography: Current Indications and Future Directions  Digital Posters2019 RSNA Case Collection Invasive Lobular CarcinomaRSNA Case Collection2021BI-RADS 4C - High suspicion for malignancyRSNA Case Collection2022Slow-growing cancerRSNA Case Collection2020 Vol. 301, No. 3 Metrics Altmetric Score PDF download" @default.
• W3200830032 created "2021-09-27" @default.
• W3200830032 creator A5069586144 @default.
• W3200830032 date "2021-09-14" @default.
• W3200830032 modified "2023-10-16" @default.
• W3200830032 title "Breast Density and Tomosynthesis" @default.
• W3200830032 cites W2151744107 @default.
• W3200830032 cites W2160408484 @default.
• W3200830032 cites W2207530733 @default.
• W3200830032 cites W2554856503 @default.
• W3200830032 cites W2946108055 @default.
• W3200830032 cites W2956095070 @default.
• W3200830032 cites W2968311158 @default.
• W3200830032 cites W3177959273 @default.
• W3200830032 doi "https://doi.org/10.1148/radiol.2021211788" @default.
• W3200830032 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/34519580" @default.
• W3200830032 hasPublicationYear "2021" @default.
• W3200830032 type Work @default.
• W3200830032 sameAs 3200830032 @default.
• W3200830032 citedByCount "0" @default.
• W3200830032 crossrefType "journal-article" @default.
• W3200830032 hasAuthorship W3200830032A5069586144 @default.
• W3200830032 hasConcept C121608353 @default.
• W3200830032 hasConcept C126322002 @default.
• W3200830032 hasConcept C126838900 @default.
• W3200830032 hasConcept C147454874 @default.
• W3200830032 hasConcept C19527891 @default.
• W3200830032 hasConcept C2780472235 @default.
• W3200830032 hasConcept C2989005 @default.
• W3200830032 hasConcept C3018951153 @default.
• W3200830032 hasConcept C530470458 @default.
• W3200830032 hasConcept C71924100 @default.
• W3200830032 hasConceptScore W3200830032C121608353 @default.
• W3200830032 hasConceptScore W3200830032C126322002 @default.
• W3200830032 hasConceptScore W3200830032C126838900 @default.
• W3200830032 hasConceptScore W3200830032C147454874 @default.
• W3200830032 hasConceptScore W3200830032C19527891 @default.
• W3200830032 hasConceptScore W3200830032C2780472235 @default.
• W3200830032 hasConceptScore W3200830032C2989005 @default.
• W3200830032 hasConceptScore W3200830032C3018951153 @default.
• W3200830032 hasConceptScore W3200830032C530470458 @default.
• W3200830032 hasConceptScore W3200830032C71924100 @default.
• W3200830032 hasLocation W32008300321 @default.
• W3200830032 hasOpenAccess W3200830032 @default.
• W3200830032 hasPrimaryLocation W32008300321 @default.
• W3200830032 hasRelatedWork W2011307685 @default.
• W3200830032 hasRelatedWork W2023780630 @default.
• W3200830032 hasRelatedWork W2043014171 @default.
• W3200830032 hasRelatedWork W2538424924 @default.
• W3200830032 hasRelatedWork W2618309483 @default.
• W3200830032 hasRelatedWork W2775569524 @default.
• W3200830032 hasRelatedWork W2802495543 @default.
• W3200830032 hasRelatedWork W3024990341 @default.
• W3200830032 hasRelatedWork W4283824386 @default.
• W3200830032 hasRelatedWork W4312982758 @default.
• W3200830032 isParatext "false" @default.
• W3200830032 isRetracted "false" @default.
• W3200830032 magId "3200830032" @default.
• W3200830032 workType "article" @default.