The BI-RADS (Breast Imaging Reporting and Data System), developed by the American College of Radiology (ACR) in the early 1990s, is the international standard for the interpretation and reporting of breast images. Initially developed for mammography, it incorporated ultrasound and magnetic resonance imaging (MRI) in subsequent editions. The fifth edition, published in 2013, brought together all three modules in a single volume for the first time.
However, over time limitations accumulated: persistent interobserver variability, incomplete incorporation of emerging technologies such as digital tomosynthesis (DBT), contrast-enhanced mammography (CEM), and abbreviated MRI protocols, and increasing regulatory demands regarding breast density reporting. This prompted the publication of the BI-RADS v2025 manual, the sixth edition of the system.
One of the first changes that stands out is that the document is no longer called an «Atlas» but rather a «Manual.» This is not a minor detail: it reflects the transition from an illustrated lexicon reference to a structured clinical practice guide that integrates reporting logic, management recommendations, and auditing standards into a single, everyday tool.
Another key change is the harmonization of the report structure across all modalities. Previous versions varied in the hierarchy of descriptors and the report sequence depending on the modality used, leading to inconsistencies in multimodal readings and difficulties for tumor boards. Version 2025 standardizes the report sections—clinical indication, technique, findings, category, and management recommendation—regardless of whether it is a mammogram, ultrasound, MRI, or CT scan.
Finally, this edition strengthens the audit and results monitoring framework, aligning the definitions of screening versus diagnostic studies across modalities and incorporating updated performance benchmarks. This consolidates the role of BI-RADS not only as a reporting lexicon but also as a tool for continuous quality improvement.
We are sharing with you two articles that summarize the differences between the previous and new versions of BI-RADS:
ACR BI-RADS® v2025 Manual – What’s New?
Recent advances in breast imaging reporting: a review of the BI-RADS® sixth edition
New features in the new BI-RADS
New developments in mammography
In mammography, the most relevant changes involve three areas: breast density, tomosynthesis, and calcifications.
Regarding breast density, the system maintains the four known categories but emphasizes standardized wording and greater alignment between the professional report and communication with the patient. This responds to regulatory requirements that have proliferated in various countries, rather than to new biological assumptions.
Regarding digital tomosynthesis (DBT), the definition of «mass» has been updated. In conventional two-dimensional mammography, confirming the three-dimensionality of a lesion required visualizing it in orthogonal projections. However, multiple studies have demonstrated that DBT allows for the characterization of lesions visible in a single tomosynthesis slice with sufficient confidence. Version 2025 formally incorporates this criterion, resulting in a reduction of unnecessary recalls and additional studies.
Regarding calcifications, the terminology is simplified by eliminating etiological descriptors of low reproducibility and prioritizing morphological ones that have greater support in the literature and greater consistency among readers.
New developments in ultrasound
Breast ultrasound receives three significant updates in this edition.
The first is the incorporation of refined tissue composition descriptors (background echotexture). Although previous studies had already demonstrated that background echotexture influences lesion detection and diagnostic performance, earlier editions did not provide sufficiently standardized guidance in this regard. Version 2025 closes that gap.
The second new feature—and perhaps the most conceptually relevant—is the formal recognition of non-mass lesions on ultrasound. The literature has long described infiltrative or poorly defined sonographic abnormalities that do not meet strict mass criteria but are associated with malignancy, particularly ductal carcinoma in situ and invasive lobular carcinoma. The lack of standardized terminology for these lesions led to inconsistent reports; version 2025 incorporates them into the lexicon and harmonizes their description with existing MRI terminology.
The third update concerns lymph node assessment, which now has its own dedicated framework encompassing intramammary, axillary, and regional lymph nodes. The emphasis shifts from lymph node size to morphological parameters—cortical thickness, integrity of the fat hilum, vascularity—which have demonstrated greater predictive value for malignancy.
New developments in magnetic resonance imaging (MRI)
MRI is the modality that receives the most conceptual changes in this edition.
The most impactful change is the elimination of the descriptor «focus.» Multiple studies had documented its poor interobserver reproducibility and the inconsistency in associated clinical management. With current advances in spatial resolution, most enhancement foci can be characterized as small masses or as non-mass enhancement, rendering the term virtually obsolete. Its elimination improves reporting clarity and reduces ambiguity.
The T2 signal is also incorporated as a mass subdescriptor. Accumulated evidence indicates that uniformly hyperintense T2 masses with benign morphology have a very low probability of malignancy. Version 2025 includes this characteristic as supporting information—not as a determinant of the final category—achieving a balance between clinical utility and interpretive caution.
Another new feature is the inclusion of peritumoral edema as an associated finding. Previous research has linked extensive peritumoral T2 hyperintensity with aggressive tumor biology, lymphovascular invasion, and increased lymph node involvement. Its inclusion allows radiologists to convey biologically relevant information without altering standardized assessment categories.
Finally, version 2025 formally recognizes abbreviated MRI protocols. Several prospective studies have demonstrated that these protocols maintain cancer detection rates comparable to full protocols, with a substantial reduction in acquisition and interpretation time. The manual establishes minimum acquisition requirements, legitimizing a practice already adopted in multiple centers.
Contrast-enhanced mammography (CEM): a new central modality
One of the most significant changes in version 2025 is the elevation of EMS to the primary modality in the manual, no longer as an accessory supplement. This decision is based on a growing body of evidence: meta-analyses and cohort studies have demonstrated that EMS offers diagnostic performance comparable to MRI in the detection and staging of lesions in certain clinical contexts.
Version 2025 integrates enhancement descriptors specific to CEM, harmonized with the NMR lexicon where appropriate, and incorporates it into the same reporting, categorization, and auditing framework as the other modalities. This is especially relevant in environments where access to NMR is limited, since CEM can be a valid alternative and now has explicit regulatory support.
Conclusion
The BI-RADS® v2025 manual represents a comprehensive and careful evolution of the breast imaging reporting system. By harmonizing the structure across modalities, eliminating ambiguous descriptors, formally incorporating tomosynthesis and abbreviated MRI protocols, and elevating CEM to a core module, the sixth edition strengthens the role of BI-RADS both as a clinical tool and as a quality assurance instrument. Its understanding and adoption are essential for a consistent, reproducible, and patient-centered contemporary breast imaging practice.
This material was automatically translated from medicosradiologos.com.ar
For any comments or suggestions: radiologyzones@gmail.com
