Breast Imaging Reporting and Data System v2025: Key Updates in Mammography Authors: Bong Joo Kang Published: 2026 ABSTRACT AbstractIn December 2025, the American College of Radiology released the 6th edition of the Breast Imaging Reporting and Data System (BI-RADS® v2025). This review aimed to help radiologists adapt to the updated guidelines by summarizing major changes in mammography, including descriptions of breast-density classification, terminology updates, and revised interpretation/reporting criteria. The mammography section of the BI-RADS® v2025 manual was comparatively analyzed against the 5th edition (2013), with an additional review of the ‘What’s New’ summary and the terminology lexicon to identify key updates in descriptors and reporting structure. Aligning with the 2024 U.S. Food and Drug Administration (FDA) breast-density notification requirements, singular wording has been permitted for unilateral examinations. Digital breast tomosynthesis (DBT) has been fully integrated into the mammography framework, allowing mass definition even on a single-projection image. Regarding terminology, the ‘lobulated’ shape descriptor has been reintroduced, while the ‘microlobulated’ margin descriptor has been removed. Regarding calcification descriptors, ‘popcorn’ and ‘dystrophic’ have been consolidated under ‘coarse,’ and the term ‘developing asymmetry’ has been retired. Lymph-node reporting has been unified into a single category; several subsections previously included under ‘Associated Features’ have been renamed as ‘Secondary Findings’ and have been conceptually redefined. In conclusion, BI-RADS® v2025 incorporates advances in DBT technology and evolving regulatory requirements. Radiologists should become familiar with these updated terms and structural changes to ensure accurate communication and standardized interpretation and reporting.Keywords: Mammography, Breast Neoplasms, BI-RADS, Terminology, Breast Density, Diagnosis INTRODUCTION Since its introduction in 1993, the American College of Radiology (ACR) Breast Imaging Reporting and Data System (BI-RADS®) has become the global standard for breast imaging interpretation and reporting (1). In December 2025, the ACR released its highly anticipated 6th edition (v2025), marking a significant update from the 5th edition in 2013 (2). Over the past decade, technological advancements, particularly the widespread adoption of digital breast tomosynthesis (DBT) and contrast-enhanced mammography (CEM), have transformed breast-cancer screening and diagnosis (3,4,5). Furthermore, regulatory landscapes have evolved, notably with the U.S. Food and Drug Administration (FDA) updating mammography quality standards for breast-density reporting by 2024 (6). This review summarizes critical updates in the Mammography section of BI-RADS® v2025 (7). The new edition aims to harmonize reporting with modern clinical practice, simplify lexicon descriptors to reduce ambiguity, and integrate findings from newer modalities, such as DBT. Key changes include terminology revision in breast-density reporting, including the allowance for singular phrasing in appropriate settings, redefined mass and calcification descriptors, and structural changes to report organization (8,9). Understanding these updates is crucial for radiologists to maintain high-quality standardized reporting that facilitates effective patient management. GENERAL UPDATES IN BI-RADS® v2025 (Table 1) Table 1Key Updates in Mammography in the ACR BI-RADS® v2025 Compared with the 2013 BI-RADS® Atlas (5th Edition)FindingsTermsSection2013 BI-RADS® Atlas (5th Edition)ACR BI-RADS® v2025ChangesMassDefinitionGenerally required visibility on two projectionsCan be defined on a single projection if DBT criteria are metDefinition expanded with DBT integrationShapeOval, round, irregularOval, round, irregular, lobulated‘Lobulated’ reinstatedMarginMicrolobulated includedMicrolobulated removedDescriptor removedCalcificationCoarse/popcorn-like/dystrophicPopcorn-like and dystrophic described separatelyUnified under coarseTerms consolidatedMilk of calciumMilk of calciumLayeringTerminology changedPunctatePunctate usedReplaced by roundDescriptor simplifiedVascular calcificationBenign descriptorBenign, with note on possible cardiovascular relevanceClinical implication addedTerminologyDeveloping asymmetryUsed as a descriptorDiscontinuedTerm removedMultiple dilated ductsNot specifically listedAdded as a benign descriptorNew term addedACR = American College of Radiology, BI-RADS = Breast Imaging Reporting and Data System, DBT = digital breast tomosynthesis Key Updates in Mammography in the ACR BI-RADS® v2025 Compared with the 2013 BI-RADS® Atlas (5th Edition) ACR = American College of Radiology, BI-RADS = Breast Imaging Reporting and Data System, DBT = digital breast tomosynthesis The structure of mammography reports has been standardized to ensure consistency across institutions (2,7). The recommended report organization strictly follows the sequence: 1) Indication, 2) Comparison, 3) Technique, 4) Breast density, 5) Findings, 6) Assessment, and 7) Management. This structured approach ensures that essential clinical information and technical details are not overlooked. In previous editions, breast-imaging indications were primarily described as screening and diagnostic. In the current edition, BI-RADS® introduces structured categories for major clinical indications for breast imaging, together with associated optional subcategory indications and relevant clinical histories. These structured indications are recommended for use in reports across all imaging modalities to improve reporting clarity and facilitate outcome monitoring, including the method of cancer detection. The major indication categories included asymptomatic screening, diagnostic workup, and diagnosis of current breast cancer. Optional subcategories and relevant clinical histories may be included when appropriate to provide additional clinical context and nuances for reporting. Another update is that BI-RADS® v2025 explicitly adds a ‘Technique’ section to the report structure, which may include digital mammography (DM), DBT, synthetic mammography (SM), specific diagnostic views, and the use of CAD or AI, and the full integration of DBT into the lexicon (3,5,10). Previously treated as supplemental technology, DBT findings are now foundational for mammographic interpretation. The manual explicitly addresses the technical descriptions of both standard DM and SM, acknowledging that SM is now a standard component of DBT protocols (11,12). BREAST-DENSITY CLASSIFICATION Breast-density reporting has undergone essential revisions in alignment with the FDA’s 2024 final rule on mammography quality standards (6). In BI-RADS® v2025, the terminology has shifted from ‘breast composition’ to ‘breast density,’ reflecting the current reporting language. The four categories of breast density (a, b, c, and d) remain fundamentally similar in concept but have been updated in definition, phrasing, and illustration to ensure clarity and regulatory compliance (13,14). FDA ALTERNATIVE STANDARD #26 In July 2025, the FDA approved Alternative Standard #26 to amend the original Mammography Quality Standards Act (MQSA) standard (21 CFR 900.12[c][2]) (15). This allows the interpreting physician to provide an overall assessment of breast density phrased in the singular for unilateral mammograms. Singular phrasing is intended for reports of unilateral mammograms, such as postmastectomy surveillance or single-breast diagnostic studies. The approved terminology for singular phrasing is as follows: 1) The breast is almost entirely fatty2) There are scattered areas of fibroglandular density3) The breast is heterogeneously dense, which may obscure small masses4) The breast is extremely dense, which lowers the sensitivity of mammography 1) The breast is almost entirely fatty 2) There are scattered areas of fibroglandular density 3) The breast is heterogeneously dense, which may obscure small masses 4) The breast is extremely dense, which lowers the sensitivity of mammography This addresses the long-standing grammatical awkwardness when reporting on a single breast in postmastectomy or unilateral diagnostic examinations, where plural terminology was technically inaccurate (16). VISUAL REFERENCES To support these definitions, the new manual provides updated reference images that include both standard DM and SM (2,17). This addition recognizes that the appearance of the fibroglandular tissue can vary slightly between standard 2D acquisitions and synthetic reconstructions from DBT, providing a more comprehensive reference for modern reading workflows (Fig. 1) (18,19,20). Images of breast-density classification. A. Representative images of the four density categories using standard digital mammography. B. Corresponding examples using SM of DBT. The inclusion of SM ensures that reference standards match modern DBT workflows. DBT = digital breast tomosynthesis, SM = synthetic mammography MASSES (Table 2) Table 2Changes in Breast-Finding Descriptions (Mass and Calcifications)FindingsTerms2013 BI-RADS® Atlas (5th Edition)ACR BI-RADS® v2025ChangesA. Masses1. Shapea. Ovala. OvalAddedb. Lobulatedb. Roundc. Roundc. Irregulard. Irregular2. Margina. Circumscribeda. CircumscribedRemovedb. Obscuredb. Obscuredc. Microlobulatedd. Indistictc. Indisticte. Spiculatedd. Spiculated3. Densitya. Fat-containinga. Fat-containingb. Low densityb. Low densityc. Equal densityc. Equal densityd. High densityd. High densityB. Calcifications1. Typically benigna. Skina. Skinb. Vascularb. VascularBenign, but noted association with cardiovascular disease riskc. Coarse or “popcorn-like”c. Coarse‘Popcorn-like’ removed; use ‘coarse’d. Large rod-liked. Large rod-likee. Round/punctatee. Round‘Punctate’ removed; use ‘Round’f. Rimf. Rimg. DystrophicCombined into single descriptor ‘Coarse’h. Milk of calciumg. LayeringReplaced by ‘Layering’i. Sutureh. Suture2. Suspicious morphologya. Amorphousa. Amorphousb. Coarse heterogenousb. Coarse heterogenousc. Fine pleomorphicc. Fine pleomorphicd. Fine linear or fine linear-brancingd. Fine linear or fine linear-brancing3. Distributiona. Diffusea. Diffuseb. Regionalb. Regionalc. Groupdc. Groupdd. Lineard. Lineare. Segmentale. SegmentalACR = American College of Radiology, BI-RADS = Breast Imaging Reporting and Data System Changes in Breast-Finding Descriptions (Mass and Calcifications) ACR = American College of Radiology, BI-RADS = Breast Imaging Reporting and Data System The definition and description of masses have been refined to reflect the superior conspicuity provided by tomosynthesis and to address the historical confusion between masses and asymmetries (2,7). MASSES VS. ASYMMETRIES All types of asymmetries have different contours than true masses and often lack the conspicuity of masses (8). Asymmetries appear similar to other discrete areas of benign fibroglandular tissue, except that they are unilateral with no mirror-image correlate in the opposite breast. Asymmetry demonstrates a concave contour and is usually interspersed with fat, whereas a mass demonstrates a completely or partially convex contour and (when radiodense) appears denser in the center than at the periphery (21,22). DEFINITION OF A MASS Historically, a mass (by definition) had to be visualized on two orthogonal mammographic projections; if visualized in only one projection, it was classified as an asymmetry (1,23). In BI-RADS® v2025, acknowledging the 3D nature of DBT, if a one-view finding demonstrates the characteristic features of a mass (e.g., convex contour and central radiodensity), it may be appropriate to classify it as a mass (Fig. 2, Table 2) (2,5,10). True masses may be observed in only one view owing to masking by overlapping fibroglandular tissue or exclusion from the image field. One-view-only masses may be more apparent on DBT (11,24). This change significantly affects the recall and assessment of findings seen in only one view (e.g., craniocaudal [CC] or mediolateral oblique [MLO]) during DBT screening. Mass identification on single-view DBT. A 63-year-old woman with a screening finding seen only on the mediolateral oblique view. A. A 2D synthesized image shows a subtle opacity. B. Sequential DBT slices clearly demonstrate a discrete mass with spiculated margins (arrow), fulfilling the new v2025 definition of a mass visible on a single projection. DBT = digital breast tomosynthesis SHAPE - LOBULATED REINSTATED The shape descriptor ‘lobulated’ (one or more shallow undulations) has been reinstated, with the verbiage slightly changed from the prior term descriptor ‘lobular’ to avoid overlap with the cancer-type descriptor in lobular carcinoma (23,25). Based on expert consensus, a lobulated shape was felt to convey a more suspicious appearance than a strictly oval shape and was therefore returned to the lexicon to allow a more nuanced shape description (Fig. 3) (22,26,27). Reinstated lobulated shape. A, B. Examples of masses that fit the reinstated lobulated shape descriptor. These lesions exhibit undulating contours that are distinct from the smooth curve of an oval mass but lack the random jagged appearance of an irregular shape. MARGIN - MICROLOBULATED REMOVED The margin descriptor ‘microlobulated’ (many small undulations) has been removed to avoid overlapping terminology between shape and margin (2,7). A microlobulated margin should now be described as indistinct (Fig. 4). This simplification aims to reduce inter-observer variability and clarify reporting (22,26). Indistinct vs. removed microlobulated margin. A, B. Examples of masses with fine margin undulations previously termed microlobulated. Under v2025, this finding is categorized as having indistinct margins, simplifying the lexicon to reduce inter-observer variability. CALCIFICATIONS (Table 2) Calcification analysis remains the cornerstone of mammographic interpretation (8,28,29). The BI-RADS® v2025 manual also notes that calcifications may appear enhanced or altered on synthetic mammography and that pseudocalcifications may occur, which should be considered during interpretation (2,7). COARSE Among descriptors of typically benign calcifications, the term ‘popcorn like’ has been eliminated by incorporating it into the remaining descriptor category ‘coarse’ (Fig. 5A) (2,7). This consolidation shifts the lexicon away from food-based metaphors towards more descriptive morphological terms. This category also now includes dystrophic calcifications, which have been eliminated as a separate category (Fig. 5B). The clinical implication—that these are typically benign findings, often related to fibroadenomas or post-traumatic changes—remains unchanged (8). Coarse calcifications. A compilation of calcifications now grouped under the single descriptor ‘coarse’. A. Classic involuting fibroadenomas previously called popcorn-like. B. Large irregular calcifications previously termed dystrophic post-trauma. Both are now classified simply as ‘coarse.’ LAYERING The term ‘milk of calcium’ has been replaced by ‘layering’ to emphasize the morphologic appearance of the underlying process (i.e., layering or sedimentation of calcium within micro- or macro-cysts) rather than the physiologic make-up of the sediment (2,7). This change focuses on the visual appearance on the lateral view (meniscus or ‘teacup’ shape) (Fig. 6) (28). Layering calcifications. A. Craniocaudal view showing amorphous or round calcifications. B. Mediolateral oblique or true lateral view demonstrating the classic meniscus or teacup appearance. The term ‘milk of calcium’ is replaced by the descriptive term ‘layering’ in the new edition. PUNCTATE REMOVED The term ‘punctate’ has been removed as a descriptor (2). Historically, round calcifications were termed punctate when <0.5 mm and round when ≥0.5 mm. The 5th edition made punctate a subcategory of round calcifications (1). However, given the impracticality of measuring the size of individual calcifications, ‘punctate’ has been removed as a descriptor (7). Any calcification with a round shape should be characterized as ‘round’ regardless of its size. The simplification of round calcifications (elimination of the punctate subcategory) allows for more consistent reporting (22,26). They should generally be categorized as round or oval unless they display suspicious distribution patterns (8,29). VASCULAR This update explicitly notes the association between vascular calcifications and increased cardiovascular-disease risk, encouraging radiologists to consider this a reportable health marker, although it remains a benign breast finding (2). Vascular calcifications are considered typically benign findings, although currently there is no consensus on whether the presence of breast arterial calcifications should be routinely reported. ARCHITECTURAL DISTORTION (Table 3) Table 3Changes in Breast-finding Descriptions (Associated Features and Other Findings)FindingsTerms2013 BI-RADS® Atlas (5th Edition)ACR BI-RADS® v2025ChangesC. Architectural distortionD. Asymmetries1. Global asymmetry1. Global asymmetry2. Asymmetry2. Asymmetry3. Focal asymmetry3. Focal asymmetry4. Developing asymmetryRemovedE. Lymph nodesIntramammry lymph node1. Intramammarya. Normal b. Abnormal2. Axillarya. Normal b. AbnormalRelocatedF. Skin lesionsG. Dilated ductsSolitary dilated duct1. MultipleAdded2. SolitaryH. Associated features1. Skin retraction1. Skin retraction2. Nipple retraction2. Nipple retraction3. Skin thickening3. Skin thickening4. Trabecular thickening4. Trabecular thickening5. Axillary adenopathyRemoved6. Architectural distortionRemoved7. CalcificationsRemovedI. Special cases1. GynecomastiaAdded2. Implants and other forms of augmentation3. MastectomyJ. Location of finding1. Laterality1. LateralityRedirection2. Quadrant and clock face2. Location in breast3. Depth3. Depth and/or distance from nipple4. Distance from the nippleACR = American College of Radiology, BI-RADS = Breast Imaging Reporting and Data System Changes in Breast-finding Descriptions (Associated Features and Other Findings) ACR = American College of Radiology, BI-RADS = Breast Imaging Reporting and Data System Architectural distortion remains a critical finding, particularly with the high sensitivity of DBT (5,10,11,24). In v2025, architectural distortion is maintained as a primary finding type (2). However, its categorization when associated with a mass has changed. If architectural distortion is present in association with a visible mass, the mass is the primary finding, whereas the architectural distortion becomes a ‘Secondary Findings’. This clarifies the reporting hierarchy: report the mass first and the distortion next (7). ASYMMETRIES (Table 3) In earlier editions of the BI-RADS®, asymmetries were described as four entities—asymmetry, global asymmetry, focal asymmetry, and developing asymmetry—distinguished by their extent, projectional visibility, and temporal changes. Asymmetry is observed on only one projection and most often reflects the summation of normal tissue, whereas global asymmetry involves at least one quadrant without associated suspicious features and usually represents a normal variation in the fibroglandular distribution. Focal asymmetry is an area of tissue confined to less than one quadrant that is asymmetric to the corresponding location in the contralateral breast (8). It is visible and may appear similar on multiple mammographic projections (thus, a real finding rather than a superimposition of normal breast structures) but lacks the convex contour, central density, and conspicuity of a mass. The contour of focal asymmetry is concave, and interspersed fat is usually present (21,22). In the previous edition, ‘developing asymmetry’ described an asymmetry that was new, larger, or more conspicuous compared to prior examinations (1). The BI-RADS® committee decided that the term ‘developing asymmetry’ as a descriptor has been discontinued, to be consistent with the remaining lexicon where the denotation of change is not embedded in the descriptor terminology (2,7). Additionally, on occasion, the term was not fully descriptive, as it was intended to describe a new, larger, or more conspicuous focal asymmetry, but only contained the term ‘asymmetry.’ In v2025, such findings can be described simply as focal asymmetries, and if desired, a change in appearance over time can be commented upon in the comparison section of the report. Although the terminology has changed, the clinical importance of new or increasingly conspicuous asymmetry or focal asymmetry should be emphasized (29). If confirmed to be a real finding by diagnostic evaluation, biopsy should be recommended even in the absence of an ultrasound correlate, as the positive predictive value (PPV) for such a finding is in the range of 7.4%–12.8% when detected on screening and 19.7%–26.7% in the diagnostic setting (30,31,32). Radiologists should describe the finding based on its current morphological appearance (e.g., asymmetry, focal asymmetry, or mass if it meets the mass criteria) and describe the temporal change in the comparison section of the report. This enforces a strict separation between the description (morphology) and assessment (changes over time). LYMPH NODES (Table 3) Previously, lymph nodes were separated based on their location (1,23). In BI-RADS® v2025, ‘Lymph Nodes’ is the single primary finding category (2,7). Within this category, the radiologist specifies the location as a subfinding: ‘intramammary,’ ‘axillary,’ or other. This streamlines the lexicon hierarchy, grouping all lymph-node findings together regardless of their anatomical position within the field of view. DILATED DUCTS (Table 3) The lexicon has refined the assessment of dilated ducts based on newer literature that confirms prior research (2,7). SOLITARY DILATED DUCT In the 5th edition of BI-RADS®, tissue diagnosis is recommended for a solitary dilated duct unless a benign etiology is demonstrated (Fig. 7A) (1). Newer studies confirm prior research that a solitary dilated duct is a rare finding and, when not associated with suspicious imaging features (e.g., associated mass, architectural distortion, or calcifications) and occurs in asymptomatic individuals, has essentially a 0% likelihood of malignancy and can be considered benign (2,7). Solitary dilated duct. A. A solitary dilated duct (arrows) in an asymptomatic patient without associated mass or calcifications, which v2025 now permits to be assessed as a benign finding. B. A solitary dilated duct (arrows) in a symptomatic patient with bloody nipple discharge, which was considered tissue diagnosis, confirmed as sclerosing papilloma. A solitary dilated duct may be considered benign in an asymptomatic patient without other suspicious findings; however, when it is new or associated with clinical or other mammographic abnormalities, further evaluation with diagnostic imaging may be warranted to exclude malignancy (Fig. 7B). This represents a significant shift intended to reduce unnecessary recalls and workups for incidental solitary ducts seen on DBT screening, where the superior visualization of the ductal structures makes this finding more frequently detected (10,11,24). MULTIPLE DILATED DUCTS ‘Multiple dilated ducts’ have been added as a typical benign finding in the current lexicon, reflecting the benign nature of this common finding in clinical practice (2). REDEFINITION OF ASSOCIATED FEATURES AND SECONDARY FINDINGS (Table 3) Several subsections previously included under associated features have been renamed as secondary findings and have been conceptually redefined (2,7). This represents a fundamental shift in the way radiologists structure their reports. The primary finding is an abnormality that drives the BI-RADS recommendation and management (8). Secondary findings are additional abnormalities present in association with the primary finding (2). For example, there may be a suspicious mass (primary finding) for which an ultrasound-guided biopsy is recommended as well as calcifications (secondary finding) within the mass (Fig. 8). Alternatively, there may be suspicious calcifications (primary finding) for which a stereotactic biopsy is recommended as well as a subtle asymmetry (secondary finding) overlapping the calcifications (33). In contrast, associated features, when appropriate, allow for a more nuanced description of the imaging presentation of abnormalities (7). Secondary findings can include any of the primary finding types (such as masses, calcifications, asymmetries, and architectural distortions) when they occur in association with another finding deemed the primary driver of management recommendations. Items recategorized as secondary findings include. Secondary findings. A, B. Craniocaudal and mediolateral oblique views of an example illustrating the hierarchy change. A primary irregular mass (arrows) with microcalcifications is associated with skin retraction (arrowheads) and axillary adenopathy. The mass is reported as the primary finding, the microcalcifications and axillary adenopathy are categorized as secondary findings, while the skin retraction (arrowheads) is categorized as associated features. • Architectural distortion: When seen with a mass or calcifications (2)• Axillary adenopathy: When associated with a breast lesion (2,7)• Calcifications: When associated with a mass or architectural distortion (2,33) • Architectural distortion: When seen with a mass or calcifications (2) • Axillary adenopathy: When associated with a breast lesion (2,7) • Calcifications: When associated with a mass or architectural distortion (2,33) This distinction helps structure the report logically: identify the dominant lesion (primary finding) that drives the BI-RADS assessment and management and list the supporting evidence or concurrent abnormalities (secondary findings) (8,21). The use of primary and secondary findings, as well as associated features, when appropriate, allows for a more nuanced description of the imaging presentation of abnormalities. In the current edition, ‘Associated Features’ refer specifically to breast changes that occur as a consequence of a primary finding, including nipple retraction, skin retraction, trabecular thickening, and skin thickening (7). SPECIAL CASES (Table 3) ‘Special Cases’ have been reintroduced in this edition to capture specific clinical scenarios or findings that do not fit neatly into morphological descriptors (2,7). This category includes the following: • Gynecomastia• Implants and other forms of augmentation (including injections)• Mastectomy • Gynecomastia • Implants and other forms of augmentation (including injections) • Mastectomy Grouping these under Special Cases allows for the standardized reporting of post-surgical and non-cancerous conditions that require specific management or technical adjustments (23). LOCATION DESCRIPTION (Table 3) Describing the lesion location has become more flexible in the current version (2,7). Compared with the previous edition, BI-RADS® v2025 allows for a more flexible location description, particularly in the context of DBT and other contemporary modalities, emphasizing depth and/or distance from the nipple as practical descriptors. • Quadrant or clock face notation• Depth: anterior, mid, posterior• Distance from the nipple • Quadrant or clock face notation • Depth: anterior, mid, posterior • Distance from the nipple DEPTH ON MLO VIEW A technical update for the MLO view specifies that depth (anterior/mid/posterior) should be determined using imaginary divisions that are parallel to the angle of the pectoralis major muscle rather than using vertical divisions (Fig. 9) (2,7). This method was felt to be more anatomically logical. In this way, findings that project over the axilla but are anatomically located at the posterior depth, rather than occasionally at the middle depth as could occur using the prior system of depth assignment, could be appropriately reported. This aligns the location description more accurately with the anatomical compression of the breast tissue during MLO acquisition and facilitates the correlation between mammography and other breast imaging modalities, given the differences in patient positioning (34,35). CC and MLO location depth. A, B. Schematic showing the updated method for determining lesion depth in the CC (A) and MLO (B) views. Lines dividing the breast into anterior, middle, and posterior depths are drawn parallel to the pectoralis major muscle angle rather than vertically to better reflect the anatomical compression. CC = craniocaudal, MLO = mediolateral oblique Enhanced flexibility in location description allows radiologists to choose the most appropriate method based on specific clinical scenarios and institutional preferences while maintaining standardization through clear guidelines (8). ASSESSMENT CATEGORIES Although the overall assessment category framework is largely similar to that of the 5th edition, BI-RADS® v2025 more clearly standardizes report organization and clarifies the appropriate application of the selected categories (1,2,7,8). CATEGORY 4 (SUSPICIOUS) SUBDIVISIONS Category 4 findings have a likelihood of malignancy of 2%–95% and encompass most findings prompting breast interventional procedures (30). The subdivision into 4A, 4B, and 4C has been clarified (2). In a retrospective review of diagnostic mammography data submitted to the National Mammography Database (NMD) comprising 125,447 examinations from 41,841 women, including 44,718 examinations with category 4 assessments, the PPV3 were 7.6% for category 4A, 22% for 4B, and 69.3% for 4C (36). While subcategory designation cannot replace the overall Category 4 designation per FDA regulations (6), it may be used in addition to providing nuanced communication regarding the likelihood of malignancy (31,32). CATEGORY 1 (Negative) VS. CATEGORY 2 (BENIGN) An important clarification addresses the distinction between these categories (2,7). Both indicate no mammographic evidence of malignancy and have the same management recommendations (routine screening) (30,31). Category 1 should be used when no specific benign findings are described in the report, whereas category 2 should be used when at least one benign finding is described. This difference is related to whether benign findings are described in the report and not whether such findings are visible on examination (8). CATEGORY 3 (PROBABLY BENIGN) BI-RADS® v2025 more clearly specifies three mammographic findings that may be assessed as category 3 after complete diagnostic evaluation: a solitary non-calcified circumscribed solid mass, focal asymmetry without sonographic correlate, and a solitary group of round calcifications. These findings may be managed with short-interval imaging follow-up rather than immediate tissue diagnosis, provided that the diagnostic workup is complete and no suspicious associated features are present. The manual also emphasizes that a category 3 assessment should not be rendered directly from screening examination findings but only after diagnostic evaluation has confirmed a probably benign appearance. For category 3 assessments, the initial short-term follow-up interval is usually 6 months, followed by a second follow-up at 6 months after the initial short-term study, and continued surveillance at 12-month intervals to establish long-term stability of 2 years (1,2,8). If, at any short-interval or surveillance mammography examination, the probably benign finding decreases or resolves, the appropriate assessment is category 2 or 1. CATEGORY 6 (KNOWN MALIGNANCY) This category is used when breast imaging is performed in an individual with a known malignancy during neoadjuvant treatment and prior to definitive local treatment, typically surgical excision (2). When further breast imaging is performed prior to the definitive treatment of a biopsy-proven malignancy and additional suspicious findings other than the known cancer are demonstrated, a category 6 assessment may not be appropriate (7). The correct approach to reporting in this scenario is to provide the category 6 assessment with a concordant management recommendation for known cancers but to follow this with a sentence recommending management appropriate for the finding(s) not covered by the overall assessment. ASSESSMENT HIERARCHY When more than one type of diagnostic examination is performed on the same day, the overall assessment and management recommendations should be based on the higher actionable finding (2,7). The hierarchy of decreasing abnormalities is as follows: categories 5, 4, 0, 6, 3, 2, and 1. For example, if a diagnostic mammogram identifies a BI-RADS category 4 finding but a subsequent same-day ultrasound reveals a suspicious mass requiring biopsy (category 5), the final report should provide an overall assessment of BI-RADS category 5 (8,35). REPORT ORGANIZATION The structure of mammography reports has been standardized to ensure consistency across institutions and to facilitate clear communication (2,7,8). The recommended report organization strictly follows the following sequence: 1) Indication: The clinical reason for the examination2) Comparison: Reference to prior examinations3) Technique: Technical details of the current study4) Breast Density: Assessment using the four-category system (6,13,15)5) Findings: Detailed description of any abnormalities6) Assessment: BI-RADS category assignment (1,2)7) Management: Specific recommendations for follow-up or intervention (30,31) 1) Indication: The clinical reason for the examination 2) Comparison: Reference to prior examinations 3) Technique: Technical details of the current study 4) Breast Density: Assessment using the four-category system (6,13,15) 5) Findings: Detailed description of any abnormalities 6) Assessment: BI-RADS category assignment (1,2) 7) Management: Specific recommendations for follow-up or intervention (30,31) ASSESSMENT-MANAGEMENT CONCORDANCE A hallmark of appropriate interpretation is concordance between the BI-RADS assessment category and management recommendations (8,21). Concordance ensures that referring clinicians and patients understand the report recommendations, which helps reduce confusion and supports the correct treatment (9). Discordant management recommendations should only be used in appropriate clinical scenarios and should be clearly documented in the report to inform the referring provider and patient (2,7). For example, a common scenario requiring discordant management involves a palpable breast abnormality with negative imaging findings (category 1 or 2) (8). Although a breast-imaging report should indicate a negative assessment, the correct approach is to provide a negative (category 1) assessment with a concordant management recommendation for routine mammography screening, followed by a sentence recommending clinical management that includes surgical consultation or tissue diagnosis, if clinically indicated (33). COMPARISON TO PREVIOUS EXAMINATIONS Comparison with prior examinations should occur when the patient has images within the system (28). The retrieval of outside images for comparison is encouraged when logistically feasible. When suspicious findings are present, comparison with previous examination(s) can be useful; however, biopsy should not be delayed awaiting the arrival of prior examinations (8,32). For findings that require prompt action, comments regarding the presence or absence of interval changes may be omitted (2). This structured approach ensures that essential clinical information and technical details are not overlooked and maintains consistency in breast imaging reporting across different practice settings (9,37). DISCUSSION The BI-RADS® v2025 update represents the maturation of breast imaging standards that reflects technological advancements and a decade of accumulated clinical experience (2,7,9). The full integration of DBT into the mass definition addresses a significant gap in the previous edition, where 3D findings had to be interpreted through a framework designed for 2D imaging (3,4,5,10). The ability to confidently identify a mass on a single DBT projection validates the spatial resolution capability of tomosynthesis and may reduce the need for additional views in diagnostic work-ups, potentially streamlining patient care and reducing radiation exposure (11,12,24,38). The full integration of DBT into BI-RADS® v2025 has implications beyond terminology: in screening practice, it may affect recall and diagnostic workflow for findings visible on only one projection, whereas in diagnostic settings it may allow the more confident characterization of single-view findings as masses when DBT features are sufficiently convincing. The simplification of calcification and margin descriptors reflects a deliberate move towards higher interobserver agreement (22,26,27). By removing ambiguous ‘in-between’ categories such as ‘microlobulated’ margins and consolidating ‘popcorn-like’ and ‘dystrophic’ calcifications into a single ‘coarse’ descriptor, the lexicon encourages decisive characterization (2,7). The elimination of food-based metaphors in favor of morphologic terms (‘layering’ instead of ‘milk of calcium’) represents a professional evolution towards more precise descriptive language (28). The removal of ‘punctate’ as a distinct category acknowledges the practical limitations of precise measurements at the submillimeter scale in routine clinical practice (22). The alignment with FDA density reporting requirements, particularly Alternative Standard #26, which allows singular phrasing for unilateral mammograms, demonstrates responsiveness to regulatory changes while addressing practical reporting challenges faced by radiologists (6,15,16). This flexibility is especially valuable for postmastectomy surveillance, where grammatically awkward plural phrasing was previously required despite the examination of only one breast (13,14,39). The discontinuation of ‘developing asymmetry’ forces a clearer distinction between morphology and temporal change, potentially reducing confusion in initial screening reports (2,7). By separating the description of what is seen (morphology) from the interpretation of how it has changed (temporal assessment), the new system encourages radiologists to first characterize the findings accurately before considering interval changes (8). This is particularly important given the clinical significance of new or enlarging focal asymmetries, which maintain their high PPV (7.4%–12.8% on screening and 19.7%–26.7% on diagnostic imaging) regardless of terminology (29,30,31,32). The reconceptualization of ‘Associated Features’ as ‘Secondary Findings’ provides a more logical reporting hierarchy (2,7). This change clarifies that radiologists should identify the primary abnormality that drives management decisions and then systematically document concurrent abnormalities (8,21,33). This approach aligns with how clinicians process radiological information and may improve communication regarding complex multifocal or multicentric disease (40,41). The refinement of dilated duct assessment, particularly the acceptance of solitary dilated ducts as benign findings in asymptomatic patients without suspicious features, reflects evidence-based practice evolution (2,7). This change has practical significance in the DBT era, wherein the enhanced visualization of ductal structures has made such findings more common (10,11,24). By providing clear guidance to avoid unnecessary workup of these benign findings, the updated manual helps reduce patient anxiety and healthcare costs, while maintaining vigilance for truly concerning the findings (30,31,32). The standardization of report organization into seven mandatory components (Indication, Comparison, Technique, Breast Density, Findings, Assessment, and Management) ensures comprehensive documentation, while facilitating quality improvement initiatives and clinical audits (9,36,37). The explicit emphasis on assessment-management concordance, with detailed guidance for handling discordant clinical scenarios, provides a practical framework for complex situations such as palpable masses with negative imaging or incidental findings during cancer surveillance (2,21,33). CONCLUSION Radiologists must familiarize themselves with these lexicon changes, structural updates, and clarified assessment category guidelines to ensure regulatory compliance (6,15), maintain high-quality standardized reporting (31,32,36), and facilitate effective clinical communication (9,21,37). These updates balance technological advancements with practical clinical applications (3,4,5,10,11,12,24,38,42,43,44), ultimately improving patient care through clearer and more consistent interpretation and reporting of breast imaging (9,27,45). Footnotes Conflicts of Interest: The author has no potential conflicts of interest to disclose. Funding: None Supplementary Materials Korean translation of this article is available with the Online-only Data Supplement at http://doi.org/10.3348/jksr.2026.0028. SUPPLEMENTARY MATERIAL References Associated Data Supplementary Materials SUPPLEMENTARY MATERIAL