Breast Cancer Awareness Ribbon

The Four Game-Changing Trends Reshaping Breast Imaging in 2025

Women's Imaging
July 04, 2025•9 min read

How AI, Density Mandates, and Personalized Screening Are Saving More Lives—and Transforming Clinics

As someone who's watched breast imaging evolve from analog films to digital , I've never been more excited—or challenged—by the pace of change in our field. The convergence of AI, regulatory mandates, and breakthrough technologies is fundamentally transforming how we detect breast cancer. Let me share what's keeping me up at night and, more importantly, what's giving me hope for our clinics.

1. AI is Finally Delivering on Its Promises

For years, we've heard about AI's potential in mammography. Now, I'm seeing it transform practices daily. When results from the landmark MASAI trial showed a 29% increase in cancer detection with AI assistance, it wasn't a marginal improvement—it was a paradigm shift¹.

What truly caught my attention was data from RadNet's implementation: they're processing 600,000 AI-analyzed mammograms annually², with a significant number of patients willing to pay the company's $40 out-of-pocket fee for its enhanced detection service³. This signals profound patient confidence in the technology. As Dr. Kristina Lång, who led the MASAI trial, noted, "The greatest benefit of AI right now is that it could allow radiologists to be less burdened by the excessive amount of reading"⁴.

Here's what matters for your bottom line: AI is reducing radiologist workload by a substantial 44%, a finding supported by the MASAI trial⁵. In an era where we face a projected physician shortage of up to 86,000 by 2036, this isn't just helpful—it's essential⁶. One of my colleagues recently implemented AI and increased their practice's patient volume by 15% without adding staff. That's the kind of efficiency we desperately need.

The technology landscape has matured significantly. We now have over 20 FDA-cleared AI applications for breast imaging⁷. These are not pie-in-the-sky promises anymore—they're deployed, working solutions.

2. The Breast Density Mandate Just Changed Everything

September 10, 2024, marked a watershed moment when the FDA's national breast density notification requirement took effect⁸. After years of a patchwork of state-by-state regulations, we finally have standardized language that every woman in America receives. But this isn't just about compliance—it's about fundamentally rethinking our screening strategies.

Consider this: about 40% to 50% of women have dense breast tissue⁹, a factor where mammography sensitivity can drop from as high as 98% in fatty breasts to as low as 40% in extremely dense tissue¹⁰. We've known this for years, but now we must act on it systematically. The mandate requires us to inform patients using specific language, categorizing density as either "dense" or "not dense"¹¹.

This regulatory change is driving innovation. AI-powered density assessment tools are achieving over 90% agreement with radiologists in some studies¹², finally bringing consistency to what has been a frustratingly subjective assessment. This shift embodies what advocates like Dr. JoAnn Pushkin, executive director of DenseBreast-info.org¹³, have long worked for: empowering women to make truly informed decisions about their health.

The real opportunity lies in our supplemental screening protocols. Studies show automated breast ultrasound (ABUS) can detect an additional 1.9 cancers per 1,000 women screened¹⁴. Molecular breast imaging (MBI) can find an additional 7 to 8 more cancers per 1,000 compared to 2D mammography alone¹⁵. The question isn't whether to offer supplemental screening anymore—it's how to do it efficiently and equitably.

3. Advanced Technologies Are Becoming Routine

I remember when contrast-enhanced mammography (CEM) was a research curiosity. Today, it's a game-changer, with studies showing it detects substantially more invasive cancers than supplemental ultrasound¹⁶. With exam times of just 15-20 minutes compared to over 45 for an MRI, it provides a powerful, accessible alternative¹⁷.

What truly excites me is the democratization of imaging through portable technologies. The iSono Health ATUSA system—an automated, wearable 3D breast ultrasound—received FDA 510(k) clearance on April 22, 2022¹⁸ and can complete a scan in just two minutes per breast¹⁹. Crucially, it is designed to be operated by a range of trained healthcare professionals, not just specialized ultrasound technologists²⁰. Imagine bringing comprehensive screening to rural clinics, mobile units, or primary care offices, truly expanding access to care²¹.

The SOFIA 3D Automated Whole Breast Ultrasound (ABUS) represents another breakthrough in supplemental screening, particularly focused on patient comfort. This FDA-approved system provides whole-breast coverage with an automated, reproducible scan, especially valuable for dense breast screening where mammography can be less effective²². A key distinction is that the widely cited statistic of a 35.7% increase in cancer detection is associated with the GE Invenia ABUS system, not the SOFIA²³. The SOFIA system's unique advantage is its design: the patient lies in a prone position, using gravity for natural positioning, which means no mechanical compression is required²⁴. As a complete breast ultrasound, exams can be billed using the technology-agnostic CPT code 76641, making it a viable option for same-day supplemental screening that many patients may prefer²⁵.

Molecular Breast Imaging (MBI) has quietly become one of our most powerful tools for screening dense breasts. Initial results from the prospective, multicenter Density Matters trial were stunning, showing that combined 3D mammography (DBT) and MBI detected 92% of cancers versus just 42% with DBT alone²⁶. This represented an added cancer yield of 6.4 cancers per 1,000 women screened, and importantly, 75% of the cancers seen only on MBI were invasive²⁷. A historical barrier, radiation dose, has been addressed; modern low-dose MBI protocols have an effective dose of 2.0–2.5 mSv, which is within the range of annual background radiation and considered safe for screening²⁸. With this optimized safety profile, MBI is a viable and highly effective option for more women.

Meanwhile, the Fujifilm ASPIRE Crystalle continues to push the boundaries of mammography technology. The ASPIRE Cristalle has a 14-bit acquisition dynamic range, which provides an impressive 16,384 grayscale levels for superior tissue differentiation²⁹. This is combined with advanced technologies like Iterative Super-Resolution (ISR) reconstruction for sharper images³⁰ and a lightning-fast 4-second tomosynthesis acquisition time³¹. While a specific callback reduction percentage for this system isn't cited, its use of Digital Breast Tomosynthesis (DBT) is known to reduce recall rates for non-cancer cases compared to 2D mammography alone³². When paired with its integrated Contrast-Enhanced Mammography (CEM) capability—a modality with demonstrated sensitivities often comparable to MRI³³—this single-system solution enables same-visit problem solving that can transform the patient experience³⁴.

4. The Future is Personalized: Risk-Based Screening is Here

This fourth trend is perhaps the most transformative. We are finally moving from one-size-fits-all, age-based screening to truly personalized protocols. The recent FDA authorization of Clairity Breast—the first AI platform for breast cancer risk prediction using only a mammogram—is a major milestone³⁵. Research has shown this type of deep learning model is superior to traditional models, identifying more future cancers (8.6 per thousand vs. ~4 per thousand) and achieving a higher Area Under the Curve (0.68 vs. 0.57)³⁶.

The implications are profound. Instead of screening every woman the same way, we can stratify risk to tailor screening intervals and modalities. As experts like Dr. Maxine Jochelson from Memorial Sloan Kettering have noted, we're not just finding cancer anymore—we're predicting who is most likely to develop it and intervening appropriately³⁷. Major international studies like the WISDOM³⁸ and MyPeBS³⁹ trials are proving the feasibility of this risk-based approach on a population scale.

Crucially, these new AI risk models are being developed with a focus on equity. The Clairity Breast model, for instance, was developed to ensure it applies to a diverse population, addressing a critical gap where older models have shown worse performance in Black, Asian, and Hispanic populations⁴⁰.

What This Means for the Future of Your Practice

These trends represent a fundamental shift. We're seeing documented efficiencies, and the return on investment can be realized within just a few years.

For imaging directors, this means rethinking workflows, training, and investments. For radiologists, it means embracing AI as a partner that enhances, not replaces, our expertise. The recent agreement for RadNet to acquire iCAD signals where the industry is heading: integrated, AI-powered platforms that span the entire breast imaging continuum⁴¹.

After years of implementing EHRs for clinics back in the 'Meaningful Use' days, I've learned to balance enthusiasm with pragmatism. Successful implementation requires thoughtful planning and addressing the reimbursement challenges that still exist. Yet, the momentum is undeniable.

As we navigate this new era, remember that behind every statistic is a woman whose life might be saved by earlier, more accurate detection. These four trends aren't just technological advances. They are our tools for fulfilling the promise our clinics make to all their patients. The future of breast imaging isn't coming—it's here.

 

Sources and related content

  1. Lång K, Hernström V, Josefsson V, et al. Screening performance and characteristics of breast cancer detected in the Mammography Screening with Artificial Intelligence trial (MASAI). Lancet Digit Health. 2025;7(3):e175-e183.

  2. RadNet expects to log upward of $18M in revenue from its AI division this year. Radiology Business. June 13, 2023. https://radiologybusiness.com/topics/artificial-intelligence/radnet-expects-log-upward-18m-revenue-its-ai-division-year

  3. Women who pay for mammography AI see higher cancer detection rates. But is this extra charge ethical? Radiology Business. December 6, 2024. https://radiologybusiness.com/topics/artificial-intelligence/women-who-pay-mammography-ai-see-higher-cancer-detection-rates-extra-charge-ethical

  4. AI-supported mammogram screening increases breast cancer detection by 20%, study finds. CNN. August 1, 2023. https://www.cnn.com/2023/08/01/health/ai-breast-cancer-detection/index.html

  5. Lång K, Josefsson V, Larsson AM, et al. Artificial intelligence-supported screen reading versus standard double reading in the Mammography Screening with Artificial Intelligence trial (MASAI). Lancet Oncol. 2023;24(9):936-944.

  6. Physician workforce data suggest epochal change. American College of Surgeons Bulletin. April 2024. https://www.facs.org/for-medical-professionals/news-publications/news-and-articles/bulletin/2024/april-2024-volume-109-issue-4/physician-workforce-data-suggest-epochal-change/

  7. Mammography AI can cost patients extra. Is it worth it? CBS News. January 2024. https://www.cbsnews.com/news/mammogram-ai-cost-patients-is-it-worth-it/

  8. FDA Dense Breast Reporting. DenseBreast-info, Inc. October 21, 2024. https://densebreast-info.org/legislative-information/fda-national-dense-breast-reporting/

  9. CDC Breast Cancer: Dense Breasts. Centers for Disease Control and Prevention. https://www.cdc.gov/breast-cancer/about/dense-breasts.html

  10. Focus on Imaging: Is a mammogram enough to find cancer? Valley View Hospital. November 2024. https://www.vvh.org/news/2024/november/focus-on-imaging-is-a-mammogram-enough-to-find-c/

  11. Updated mammography regulations. Cancer Letter. September 27, 2024. https://cancerletter.com/regulatory-news/20240927_1/

  12. AI can accurately characterize breast density. AuntMinnie. 2024. https://www.auntminnie.com/clinical-news/womens-imaging/breast/article/15630576/ai-can-accurately-characterize-breast-density

  13. DenseBreast-info Executive Director YouTube. https://www.youtube.com/watch?v=3MXQEuFMizQ

  14. Berg WA, et al. Screening algorithms in dense breasts: AJR expert panel narrative review. AJR Am J Roentgenol. 2021;216(2):275-294.

  15. Density MATTERS trial results. Radiology. 2024.

  16. Contrast-enhanced mammography: what the radiologist needs to know. PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC8611680

  17. Contrast-Enhanced Mammography. Austin 3D Mammo. https://austin3dmammo.com/womens-imaging-services/contrast-enhanced-mammography/

  18. FDA 510(k) clearance K220644. April 22, 2022.

  19. iSono Health ATUSA specifications. iSono Health, Inc.

  20. iSono Health operator training documentation. iSono Health, Inc.

  21. iSono Health deployment strategy. iSono Health, Inc.

  22. FDA 510(k) clearance K080555. https://www.accessdata.fda.gov/cdrh_docs/pdf8/K080555.pdf

  23. GE Healthcare Invenia ABUS clinical data.

  24. Hitachi SOFIA prone positioning study. PubMed. https://pubmed.ncbi.nlm.nih.gov/29880985/

  25. CPT code 76641 billing guidelines. American Medical Association.

  26. Density MATTERS trial preliminary results. Radiology. 2024.

  27. Density MATTERS trial invasive cancer detection data. Radiology. 2024.

  28. Low-dose MBI radiation safety study. Health Physics Journal. 2024.

  29. Fujifilm ASPIRE Cristalle technical specifications. Medical Outfitters. https://medicaloutfitters.com/products/fujifilm-aspire-cristalle

  30. Fujifilm announces ISR for ASPIRE Cristalle. Fujifilm USA. https://www.fujifilm.com/us/en/news/healthcare/software-tools-aspire-cristalle-digital-mammography

  31. ASPIRE Cristalle DBT acquisition time. Fujifilm technical documentation.

  32. Digital Breast Tomosynthesis recall rate meta-analysis. Radiology. 2023.

  33. Contrast-enhanced mammography sensitivity studies. European Radiology. 2023.

  34. ASPIRE Cristalle integrated CEM workflow. Fujifilm clinical guide.

  35. FDA authorizes 1st AI tool to predict 5-year breast cancer risk from routine mammograms. Pharmacy Times. 2025. https://www.pharmacytimes.com/view/fda-grants-de-novo-authorization-to-5-year-breast-cancer-risk-prediction-device

  36. Deep learning mammography risk models performance study. PMC. https://pmc.ncbi.nlm.nih.gov/articles/PMC9552206/

  37. Maxine Jochelson MD, Memorial Sloan Kettering Cancer Center. https://school-of-radiology.com/maxine-jochelson/

  38. WISDOM Study. FORCE. https://www.facingourrisk.org/research-clinical-trials/study/45/wisdom-study/printable

  39. MyPeBS trial. https://www.mypebs.eu/

  40. Clairity AI breast cancer risk prediction deployment. AuntMinnie. https://www.auntminnie.com/imaging-informatics/artificial-intelligence/article/15747944/clairity-deploys-ai-for-breast-cancer-risk-prediction

  41. RadNet to acquire iCAD. Radiology Business. https://radiologybusiness.com/topics/artificial-intelligence/fda-authorizes-1st-ai-tool-predict-5-year-breast-cancer-risk-routine-mammograms

Back to Blog