Оцінка категорії BI-RADS мамографічного дослідження — від неповної оцінки (0) до підтвердженої малігнізації (6)
Every mammography report ends with a Breast Imaging Reporting and Data System (BI-RADS) final assessment category. Category 0 is not a diagnosis at all — it is a statement that the current images are insufficient to render a final assessment, and additional imaging is mandatory before the case can be closed. Category 1 is the true negative baseline: a mammogram with no masses, no suspicious calcifications, and no architectural distortion.
Category 0 is assigned when the radiologist cannot render a final assessment from the current study alone. This is overwhelmingly a screening-mammogram category — it is rarely appropriate on a diagnostic workup, where additional imaging has typically already been obtained.
Common reasons for a Category 0 recall: • A density or asymmetry seen on only one view, requiring additional projections to confirm it is a true 3-D finding rather than summation artifact • A finding that needs spot compression to assess margins, or magnification to characterize calcification morphology • A palpable area not clearly explained by the mammogram, warranting targeted ultrasound • Prior comparison films unavailable — comparison to priors is often essential to distinguish a stable finding from a new one
Category 0 always triggers a specific, documented recommendation — never a vague "follow up as needed." The additional imaging is completed in a short window (days to a few weeks), after which the case is reassigned a definitive final category (usually 1, 2, or 3; occasionally 4 or 5).
Category 1 means the breasts are symmetric and no masses, architectural distortion, or suspicious calcifications are present. This is the single most common final assessment in a well-run screening population and carries an essentially zero associated malignancy risk at the time of the exam.
A Category 1 assessment does not mean the breast tissue is free of any density — it means nothing atypical was identified. Category 1 differs from Category 2 only in that Category 2 explicitly describes a benign finding (e.g., a cyst) while Category 1 describes a mammogram with nothing at all to comment on.
Management is identical for practical purposes: return to routine age-appropriate screening interval, typically annual mammography for average-risk women.
Categories 2 and 3 both describe findings that are not being biopsied — but for very different reasons. Category 2 identifies a finding the radiologist is certain is benign (a definitive diagnosis, such as a simple cyst or an involuting calcified fibroadenoma). Category 3 identifies a finding that looks probably benign, with a small but measurable malignancy likelihood, and is managed with short-interval imaging surveillance instead of biopsy.
A Category 2 assessment is a positive diagnosis of a benign entity, not merely the absence of suspicious findings. Classic Category 2 findings include simple cysts (confirmed by ultrasound), coarse "popcorn" calcifications of an involuting fibroadenoma, oil cysts from fat necrosis, vascular calcifications, and intramammary lymph nodes.
Because the diagnosis is definitive, no biopsy and no short-interval follow-up are indicated beyond routine screening. The finding is simply described in the report so that future radiologists comparing studies recognize it as a known, stable, benign entity rather than re-triggering workup each year.
Category 3 is reserved for findings with a well-established, very low (<2%) probability of malignancy based on their imaging morphology — most classically a non-calcified, circumscribed, oval solid mass, a cluster of round/punctate calcifications, or a focal asymmetry that resolves on additional views.
Rather than immediate biopsy, the accepted management is short-interval imaging follow-up: typically at 6 months, then 12 months, then 24 months from the original finding, confirming stability at each step. If the finding is unchanged through this surveillance period, it is often reclassified as Category 2 (benign, stable). Any interval growth or morphologic change prompts biopsy and reclassification to Category 4 or 5.
Category 3 is a probabilistic category, not a guarantee — it deliberately accepts a small residual risk in exchange for avoiding an unnecessary biopsy for the vast majority of patients whose finding truly is benign.
The <2% threshold that defines BI-RADS 3 is not arbitrary — it is the risk level at which the harms of biopsying every such finding (cost, discomfort, scarring, patient anxiety) are judged to outweigh the benefit of catching the rare cancer earlier than the next surveillance round would.
BI-RADS 4 is the broadest category in the classification system, spanning a range of malignancy likelihood from just above the Category 3 threshold up to the edge of Category 5. Because a single "suspicious" label would obscure enormous clinical variability, most practices sub-stratify Category 4 into 4A (low suspicion), 4B (moderate suspicion), and 4C (high suspicion) — every sub-level still warrants biopsy, but the sub-category calibrates pre-test probability for the patient and pathologist.
A finding assessed at the low end of Category 4 (e.g., a slightly indistinct mass, or amorphous calcifications) carries a materially different pre-biopsy probability of cancer than one at the high end (e.g., an irregular mass with spiculated margins that just falls short of the near-certainty of Category 5).
The 4A/4B/4C sub-classification, formalized in the ACR BI-RADS Atlas, gives referring clinicians and patients a more precise likelihood estimate for informed consent and expectation-setting before biopsy — while preserving the unambiguous rule that every Category 4 finding, regardless of sub-level, is biopsied. The sub-category is not used to defer or avoid biopsy; it calibrates the conversation and helps correlate imaging suspicion with pathology results (audit purposes).
Sub-level assignment integrates mass shape, margin, density, and associated calcifications:
• 4A — mildly suspicious features: a palpable, partially circumscribed solid mass on ultrasound; a group of indeterminate calcifications; a probable fibroadenoma in a higher-risk context • 4B — moderately suspicious: an indistinct or microlobulated mass margin; grouped amorphous or coarse heterogeneous calcifications; new architectural distortion without a clear benign cause • 4C — highly suspicious but not classic-malignant: an irregular mass with indistinct margins; fine pleomorphic or fine linear branching calcifications without the full spiculated/highly-dense appearance that would elevate it to Category 5
Radiology-pathology concordance review compares the assigned sub-level against actual biopsy outcomes across a practice's cases, functioning as a quality-assurance loop that keeps sub-level thresholds calibrated over time.
Regardless of sub-level, the pathway is the same: image-guided core-needle biopsy (stereotactic for calcifications, ultrasound-guided for masses, or MRI-guided when the finding is MRI-only). Pathology results are then correlated with the imaging appearance — a benign result that is "concordant" with a 4A assessment supports routine follow-up, whereas a benign result that is discordant with a 4C assessment prompts repeat or surgical biopsy, because the imaging suspicion was too high to accept a benign explanation at face value.
Category 5 is reserved for findings with a malignancy probability exceeding 95% based on imaging characteristics alone — classically an irregular, spiculated, high-density mass, or fine linear branching calcifications in a segmental distribution. At this level of imaging certainty, tissue diagnosis is essentially mandatory, and management pathways (surgical consultation, staging workup) are frequently initiated in parallel with — not strictly after — the confirmatory biopsy.
The classic Category 5 mass is irregular in shape, has spiculated (radiating, star-burst) margins, and is markedly hyperdense relative to surrounding fibroglandular tissue — findings that reflect the desmoplastic reaction and infiltrative growth pattern of invasive carcinoma. Calcification patterns reaching Category 5 include fine linear or fine-linear-branching ("casting") calcifications in a segmental or linear distribution, characteristic of high-grade ductal carcinoma in situ extending along a duct system.
Because these imaging features correlate so strongly with malignancy on large-scale audits, Category 5 is deliberately reserved for a narrow, high-confidence subset of findings — over-assigning it would erode its diagnostic value and unnecessarily alarm patients whose findings genuinely belong in Category 4.
A Category 5 assessment triggers prompt image-guided core-needle biopsy without the sub-stratified likelihood conversation used in Category 4 — the imaging appearance already communicates near-certainty. Multidisciplinary teams frequently begin parallel staging discussions (additional breast MRI, possible sentinel node planning, genetic counseling referral where indicated) while awaiting pathology confirmation, because delaying that planning until pathology returns would slow the overall care timeline for a group of patients where a cancer diagnosis is the overwhelmingly likely outcome.
A benign biopsy result after a Category 5 assessment is a discordant result and must not be accepted at face value — it mandates repeat sampling (often surgical excisional biopsy) because the pre-test probability of malignancy was too high for a benign explanation to be statistically plausible without a sampling error.
Category 6 is unlike every other BI-RADS category: it is not a probabilistic assessment of a new finding at all. It is applied to a lesion that has already been confirmed malignant by prior biopsy and is now being imaged for a different purpose entirely — most often to monitor response to neoadjuvant chemotherapy or to plan the extent of surgery. Category 6 is fundamentally an administrative/tracking designation layered onto the BI-RADS lexicon for continuity of imaging workflow.
Once a lesion is proven malignant by biopsy, subsequent imaging of that same lesion is no longer trying to answer "is this cancer?" — it is answering questions like "how is the tumor responding to neoadjuvant chemotherapy?" or "what is the precise extent of disease for surgical planning?" Using Categories 0–5 for this repeat imaging would be misleading, since those categories all describe the probability that an as-yet-unbiopsied finding is malignant.
Category 6 was introduced specifically to prevent this category-mismatch: it flags, unambiguously, that the finding under discussion is already a known cancer, so that anyone reading the report — radiologist, oncologist, surgeon — immediately understands the clinical context of the imaging rather than misinterpreting it as a fresh suspicious finding requiring its own new workup.
Common scenarios generating a Category 6 assessment include:
• Pre-treatment extent-of-disease mapping — mammography and/or MRI performed after core-needle biopsy confirms malignancy, to define tumor size and multifocality/multicentricity before treatment planning • Neoadjuvant chemotherapy response monitoring — serial imaging performed during a course of chemotherapy given before surgery, tracking tumor size regression (or, less commonly, progression) to inform whether the regimen is working • Post-biopsy pre-surgical planning imaging — confirming clip placement at the biopsy site and re-assessing the lesion immediately before a planned lumpectomy or mastectomy
In all of these contexts, other findings elsewhere in the same breast or the contralateral breast that are genuinely new and unbiopsied still receive their own independent Category 0–5 assessment — Category 6 applies specifically to the already-proven lesion, not to the entire study.
Category 6 is a reminder that BI-RADS is a communication system, not merely a risk score: its purpose is to make sure every report reader — regardless of specialty — instantly understands what type of clinical question the imaging is answering.