BI-RADS v2025 updates breast imaging reporting across mammography, ultrasound, magnetic resonance imaging and contrast-enhanced mammography while retaining the established link between imaging findings, assessment categories and management recommendations. A recent review in the American Journal of Roentgenology summarises the main changes and their implications for interpretation, reporting, multidisciplinary communication and audit. The revision introduces a more consistent report structure, harmonised terminology, clearer assessment rules and formal recognition of newer imaging practices. It also expands lesion localisation, tissue composition and lymph node reporting, with modality-specific refinements intended to improve reproducibility and cross-modality comparison without replacing the core principles used in previous editions.
Must Read: Clearer CT Language Improves Appendicitis Reporting
More Consistent Structure and Assessment
The revised framework standardises report organisation across breast imaging modalities. Reports now follow a shared general structure adapted to each technique, with tissue composition, technique and acquisition parameters treated as dedicated components. Comparison with previous examinations is positioned earlier, before technical details. Clinical indications are organised into screening, diagnostic work-up and evaluation of known breast cancer, with optional subcategories providing additional context. Definitions of positive and negative examinations vary according to these indication categories, affecting how results are counted during audit.
The overall assessment system remains unchanged, but several categories are clarified. Category 0 now distinguishes between the need for further imaging and the need to obtain previous examinations. It should not be used when a finding is already suspicious, as category 4 or 5 should be assigned directly. Breast MRI now uses categories 4A, 4B and 4C, aligning its risk stratification with mammography and ultrasound. This does not change management recommendations but provides a more precise expression of malignancy risk and supports performance monitoring.
Category 6 is consistently reserved for biopsy-proven malignancy before definitive local treatment, including patients receiving neoadjuvant systemic therapy, even when imaging shows a complete response. In MRI, closely adjacent findings with similar morphology may be included within category 6 when they do not alter management. The framework does not yet incorporate cryoablation or active surveillance for selected low-risk ductal carcinoma in situ because long-term outcome evidence remains insufficient.
Terminology and Localisation Across Modalities
Several cross-modality changes aim to reduce descriptive variation. The term “lobulated” returns as a shape descriptor, replacing “lobular” to avoid possible confusion with histopathology. The margin descriptor “microlobulated” has been removed from mammography, MRI and contrast-enhanced mammography, although it remains in ultrasound. Findings previously described in this way are now classified as having indistinct margins.
Lesion location follows a consistent sequence: laterality, quadrant or clock-face position, then distance from the nipple and lesion depth. Digital breast tomosynthesis reports may include the slice number to support reidentification. Ultrasound adds localisation by tissue layer, including skin, subcutaneous fat, fibroglandular tissue, retroglandular fat and the pectoralis major muscle. This structure supports follow-up, image-guided procedures and communication with surgical teams.
Lymph nodes become a distinct reporting category across modalities. Assessment gives priority to morphology rather than size alone, including cortical thickening, loss of the fatty hilum, round or irregular shape, interval change and abnormal vascularity. Regional nodal stations are classified systematically, covering intramammary, axillary, internal mammary and supraclavicular nodes. Axillary levels use the pectoralis minor muscle as an anatomical landmark. Nodes are reported as normal or abnormal according to their morphology. Suspicious findings generally lead to targeted ultrasound and image-guided biopsy when clinically appropriate.
Tissue composition also gains greater consistency. Mammography and contrast-enhanced mammography retain breast density, while MRI retains fibroglandular tissue assessment. Ultrasound introduces the glandular tissue component, a qualitative estimate of glandular tissue within fibroglandular tissue after whole-breast scanning.
Modality-Specific Changes Refine Interpretation
Mammography changes integrate digital mammography, digital breast tomosynthesis and synthetic mammography. A lesion may be called a mass when its defining features appear on one tomosynthesis projection, even without visibility on both digital mammography views. When margins are obscured on digital mammography but circumscribed on tomosynthesis, classification follows the tomosynthesis appearance. Fat within a mass does not exclude malignancy, so assessment must include other features.
Calcification terminology is simplified. “Coarse” replaces “popcorn-like” and “dystrophic”, “round” replaces “punctate” and “layering” replaces “milk of calcium”. Suspicious calcification assessment gives greater weight to morphology than distribution. The term “developing asymmetry” is removed, although a new or increasing asymmetry remains clinically important. A solitary dilated duct in an asymptomatic patient without suspicious associated findings may now be assessed as benign.
Ultrasound introduces non-mass lesions as abnormalities without definable shape or margins. Distribution, echotexture, posterior acoustic features, ductal changes and calcifications guide assessment. The solid component determines management in mixed solid and cystic lesions. New descriptors distinguish a thin, uniform echogenic pseudocapsule, usually associated with benign circumscribed lesions, from a thick, irregular echogenic rind, more often associated with malignancy and included in lesion measurements.
MRI removes “focus” from the lexicon and encourages classification of small enhancing findings as masses or focal non-mass enhancement. Thick rim enhancement, T2 signal intensity and peritumoral oedema provide additional characterisation, while morphology remains central. Contrast-enhanced mammography enters the core framework, with enhancement terminology aligned more closely with MRI.
BI-RADS v2025 preserves structured reporting, evidence-based assessment categories and the connection between imaging findings and management while adapting the system to current breast imaging practice. Its main changes standardise report organisation, clarify assessment categories, harmonise terminology and strengthen lesion, tissue and lymph node description across modalities. Mammography, ultrasound, MRI and contrast-enhanced mammography each gain specific refinements, while audit definitions and performance monitoring become more closely integrated with reporting. The practical effect will depend on consistent use, reproducibility of newer descriptors and continued validation in clinical practice, supported by education and local quality assurance.
Source: American Journal of Roentgenology
Image Credit: iStock
References:
Minichetti P, Cereser L, Versienti E et al. (2026) BI-RADS v2025: Key Updates and Implications for Breast Imaging Practice. American Journal of Roentgenology: New Articles.
