Improving Mammography Diagnostic Reliability in Viet Nam: Positioning Quality, Breast Density and Detection Performance Audits
MEASUREMENT, LEARNING & IMPROVEMENTDIAGNOSTIC SAFETY & RISK MANAGEMENT
11/6/202627 min read


Improving Mammography Diagnostic Reliability in Viet Nam: Positioning Quality, Breast Density and Detection Performance Audits
A woman undergoes routine screening mammography and receives a report stating that no suspicious abnormalities were detected. The examination was completed, the images appear clear, and the result provides reassurance.
But what if part of the posterior breast tissue was not adequately included in the images? What if a cancer was obscured by dense fibroglandular tissue? Or what if the images were technically satisfactory, but a subtle abnormality was not recognized during interpretation?
These situations may lead to similar concerns, yet they arise for different reasons. One involves image acquisition and anatomical coverage. Another reflects a limitation of mammography associated with breast density. The third concerns how the available images were interpreted. Understanding these differences matters because each requires a different approach to quality improvement.
Mammography remains central to breast cancer detection, including the identification of abnormalities that may not be apparent during clinical examination. Its value, however, depends on more than the availability of modern imaging equipment or the production of technically readable images.
The relevant breast tissue must be adequately represented, image quality must support the intended assessment, and radiologists need to recognize important findings while understanding the limitations of the examination. Healthcare organizations also need reliable ways to evaluate how their services perform across the populations they examine.
Recent international developments provide useful guidance. The American College of Radiology (ACR) released its BI-RADS v2025 Manual in December 2025, updating standardized breast imaging terminology, reporting, and outcomes monitoring. Research published in 2026 has also provided more structured approaches to assessing mammography image quality and examining the contribution of supplemental imaging in women with dense breasts.
In Viet Nam, the Ministry of Health issued Decision No. 3898/QD-BYT on December 18, 2025, promulgating national professional guidance on breast cancer screening and early detection.
Together, these developments offer an opportunity to strengthen mammography reliability through three connected activities: improving positioning and image acquisition, recognizing the effects of breast density, and auditing cancer detection performance using meaningful clinical information.
Why a Completed Mammogram May Still Be Diagnostically Limited
Mammography uses low-dose X-rays to examine breast tissue and identify abnormalities such as masses, suspicious calcifications, and architectural distortion.
Screening mammography is generally performed to detect breast cancer in people without symptoms. Diagnostic mammography, by contrast, is used to investigate a clinical concern, clarify an abnormal screening finding, or address another specific diagnostic question.
Although the imaging technology may be similar, these examinations involve different patient populations, clinical purposes, and expectations.
A mammogram may be technically inadequate because relevant tissue was excluded during positioning. Another examination may include the necessary anatomy but provide reduced visibility because of motion, artifacts, or other acquisition problems. Even when images are technically satisfactory, dense tissue can obscure a cancer that might otherwise be visible. An abnormality may also be present on the images but remain unrecognized or insufficiently characterized during interpretation.
These limitations are not interchangeable. A positioning problem may require changes in acquisition technique, practical training, or equipment assessment. Reduced visibility associated with breast density calls for appropriate recognition and communication of the method's limitations. A suspected interpretive discrepancy may warrant structured professional review.
More than one limitation can affect the same examination. Dense tissue does not eliminate the need for good positioning, and technically adequate images do not guarantee that every clinically important abnormality will be detected.
A reliable mammography service therefore needs to examine the diagnostic process from image acquisition and interpretation through reporting, further assessment, and quality review.
Positioning Quality Determines How Much Breast Tissue Can Be Examined
Standard screening mammography generally includes two projections of each breast: craniocaudal (CC) and mediolateral oblique (MLO).
These projections provide complementary views of breast anatomy. The CC view is intended to demonstrate the breast from above, including as much relevant posterior, medial, and lateral tissue as reasonably achievable. The MLO view provides an oblique perspective that helps include posterior and superior breast tissue, particularly toward the axilla.
Obtaining these views involves more than placing the breast on the image receptor. The radiologic technologist must position the patient appropriately, bring relevant tissue into the imaging field, apply suitable compression, minimize avoidable movement, and recognize anatomical or technical circumstances that may limit the examination.
Incorrect positioning can leave breast tissue outside the recorded image. For example, inadequate inclusion of posterior tissue may reduce the ability to assess an abnormality close to the chest wall. A poorly positioned MLO view may inadequately demonstrate portions of the upper outer breast or the inframammary region.
The examination may nevertheless contain the expected number of images and appear complete in the reporting system. This illustrates why image completeness must be evaluated anatomically rather than inferred from the presence of standard projections alone.
The ACR Mammography Positioning Improvement Collaborative provides a structured approach to reviewing positioning quality. Its assessment criteria address tissue inclusion, motion, nipple position, the posterior nipple line, visualization of posterior tissue, the pectoral muscle, the inframammary fold, and other features relevant to image adequacy.
The collaborative distinguishes major from minor positioning criteria and uses defined assessment methods to improve consistency across reviewers.
An important distinction should be maintained. The ACR specifies that these particular scoring criteria are intended for quality improvement within the collaborative and should not be used as the scoring requirements of the separate ACR Mammography Accreditation Program.
The collaborative also defines which examinations and images are eligible for review. For example, its positioning assessment uses conventional two-dimensional or synthetic two-dimensional images rather than tomosynthesis image stacks, with specified exclusions.
Healthcare organizations should therefore understand the scope and purpose of a quality assessment tool before adopting its criteria.
The broader principle is that positioning quality should be evaluated using clear anatomical and technical standards that can be applied consistently without losing sight of individual patient circumstances.
Recognizing When an Image Is Inadequate
An important safeguard is the ability to recognize when a mammogram does not provide sufficient anatomical coverage or image quality for its intended purpose.
For CC projections, assessment may consider the inclusion of posterior breast tissue and whether the relevant medial and lateral regions have been adequately represented. For MLO projections, reviewers may examine the depiction of the pectoral muscle, the inframammary fold, and the extent of posterior tissue included.
Comparison of the posterior nipple line between CC and MLO projections can provide additional information about positioning adequacy.
However, these criteria need to be interpreted in context. Breast size, body habitus, mobility limitations, previous surgery, implants, and other anatomical or clinical factors can affect how much tissue can reasonably be demonstrated.
A technical criterion should not be applied so rigidly that an otherwise appropriate examination is rejected simply because the patient cannot achieve the positioning expected in an uncomplicated case.
The relevant question is whether the available images adequately address the intended clinical purpose and whether any important limitations have been recognized.
Repeat imaging may be justified when significant tissue is missing or an artifact could materially affect diagnostic interpretation. However, additional exposure is not automatically appropriate for every minor positioning imperfection.
The expected diagnostic benefit should be considered alongside patient discomfort, radiation exposure, and the likelihood that repeating a view will improve the examination. This requires professional judgment supported by appropriate institutional procedures.
For healthcare organizations in Viet Nam, practical improvement may involve clear local criteria for technically inadequate images, appropriate technologist training, and timely access to professional guidance when positioning proves difficult.
The objective is not to obtain an ideal image at any cost. It is to achieve diagnostically adequate imaging under the circumstances, recognize limitations when they occur, and respond appropriately.
Compression, Image Sharpness and Radiation Dose
Appropriate breast compression is an important component of mammography.
Compression can reduce tissue thickness, limit movement, improve tissue separation, and contribute to suitable image quality with optimized radiation exposure.
The amount of compression, however, should be appropriate for the individual patient. Insufficient compression may increase tissue overlap, reduce image sharpness, or affect the exposure required to produce an adequate image. Excessive or poorly tolerated compression may cause unnecessary discomfort, contribute to movement, or discourage future participation in screening.
The objective is not to maximize compression force without regard to the patient. It is to obtain the necessary diagnostic information using a suitable technique.
Patient communication can make a meaningful difference. Explaining why compression is necessary and encouraging patients to report significant discomfort may help technologists achieve appropriate positioning while respecting individual needs.
Motion-related artifacts require similar attention. When movement causes blurring that materially limits interpretation, an additional image or another appropriate response may be necessary.
Equipment quality assurance also matters. Detector performance, automatic exposure control, image processing, display conditions, and equipment maintenance can influence the visibility of clinically relevant findings.
Radiation dose should therefore be assessed alongside image quality rather than treated as an isolated performance target. Lower radiation exposure does not necessarily mean better care if the resulting images are inadequate for diagnosis. Conversely, unnecessary exposure should be avoided when sufficient image quality can be achieved through optimized techniques.
The appropriate balance requires competent personnel, suitable technical assessment, and compliance with applicable radiation protection requirements.
What Recent Research Adds to Image Quality Assessment
A study published in Radiography in 2026 provides a useful contribution to the standardization of mammography image-quality review.
Vandendaele and colleagues conducted a scoping literature review followed by a modified Delphi process involving a multidisciplinary panel of 23 experts.
The researchers developed consensus criteria addressing image identification, artifacts, image uniformity, symmetry, contrast, radiation dose, sharpness, anatomical coverage, and other technical characteristics.
The framework includes criteria concerning posterior breast tissue, the pectoral muscle, the inframammary fold, nipple-to-pectoralis distance, and the adequacy of anatomical representation in standard projections.
The study also identified areas requiring further clarification, including how compression adequacy should be assessed and which folds or artifacts should justify repeating an image.
The resulting checklist provides a structured approach to describing mammography image quality and distinguishing good, suboptimal, and insufficient technical quality.
This is relevant because technical assessment can vary considerably between observers. Without clear definitions, one reviewer may consider an image acceptable while another regards it as inadequate. Standardized criteria can help reduce subjectivity and make quality monitoring more consistent.
However, the study did not demonstrate that adopting the checklist automatically reduces missed cancers or improves long-term patient outcomes.
Its contribution is primarily a more consistent method for evaluating technical quality. Whether a healthcare organization achieves better diagnostic performance depends on how the criteria are implemented, whether deficiencies are corrected, and whether improvements are sustained.
A checklist becomes valuable when it supports better imaging practice, not merely when additional documentation is completed.
Breast Density Creates a Different Kind of Diagnostic Limitation
Breast density describes the relative appearance and distribution of fibroglandular and fatty tissue on mammographic images.
Dense fibroglandular tissue appears relatively radiopaque, while fatty tissue is more radiolucent. Many breast cancers also appear radiopaque. As a result, dense tissue may obscure cancers that would be more conspicuous in predominantly fatty breasts. This is known as the masking effect.
Unlike inadequate positioning or motion artifact, breast density is not a technical error made by the person performing the examination. It is a characteristic of the patient's breast tissue that affects the sensitivity of mammography.
Breast density is also associated with an increased risk of developing breast cancer.
These two effects should be distinguished. One concerns the difficulty of recognizing an existing cancer on mammography. The other concerns the underlying likelihood that breast cancer may develop. Although related, they are not the same clinical question.
A negative mammogram in a woman with extremely dense breasts may provide less reassurance about the absence of an occult cancer than a negative examination in someone with predominantly fatty breasts.
However, dense tissue does not make mammography ineffective in every patient. Mammography remains valuable, including for detecting certain suspicious calcifications that may be less readily identified through other imaging methods.
The appropriate response is to recognize and communicate the limitation, not assume that mammography should automatically be abandoned.
Understanding Breast Density Under BI-RADS v2025
On December 1, 2025, the ACR released the BI-RADS v2025 Manual, extending the established Breast Imaging Reporting and Data System.
The updated manual introduces revisions across mammography, ultrasound, magnetic resonance imaging, contrast-enhanced mammography, and auditing and outcomes monitoring. For mammography, the changes reflect developments in digital imaging technologies and updated terminology intended to improve consistency in reporting and interpretation.
The four general breast-density categories remain: almost entirely fatty, scattered areas of fibroglandular density, heterogeneously dense, and extremely dense.
An important refinement in BI-RADS v2025 is the greater emphasis on the potential of fibroglandular tissue to obscure abnormalities rather than treating density as a simple estimate of the amount of glandular tissue.
Two mammograms may contain similar amounts of dense tissue but differ in how that tissue is distributed and how strongly it interferes with the visibility of a suspicious lesion.
The assessment should therefore reflect the relevant imaging criteria and masking potential rather than an arbitrary numerical estimate.
The updated manual also provides further integration of digital breast tomosynthesis (DBT), synthetic mammography, and contrast-enhanced mammography, together with refinements to descriptions of masses, calcifications, and other findings.
DBT, often described as three-dimensional mammography, provides reconstructed images that can reduce some limitations caused by overlapping breast tissue. It may improve the conspicuity of certain abnormalities compared with conventional two-dimensional imaging.
Nevertheless, tomosynthesis does not eliminate the masking effects of dense tissue or guarantee that every cancer will be detected. Its effectiveness depends on equipment, acquisition technique, patient characteristics, and interpretation practices.
The adoption of a newer imaging technology should therefore be accompanied by appropriate technical evaluation and outcome monitoring rather than an assumption that newer equipment necessarily produces superior performance in every setting.
Breast Density Should Inform Care Without Creating Unnecessary Alarm
Breast-density information should help clinicians and patients understand the strengths and limitations of mammography. It should not be presented as a diagnosis or as evidence that cancer is present.
A woman with dense breasts may benefit from consideration of other relevant factors, including age, family history, previous breast findings, individual cancer risk, and the clinical purpose of imaging.
The decision to undertake supplemental imaging should consider both expected benefits and potential disadvantages.
Breast ultrasound may detect additional cancers that are not visible on mammography. However, it can also increase false-positive findings, additional investigations, and biopsies of benign lesions.
Breast MRI has high sensitivity in appropriately selected populations, although its use involves considerations relating to availability, cost, contrast administration, and additional investigations following suspicious findings.
A systematic review and meta-analysis by Duggan and colleagues, published in the February 2026 issue of The Breast, evaluated 36 studies of supplemental imaging among women with dense breasts.
Compared with mammography alone, supplemental MRI was associated with approximately 18.92 additional cancers detected per 1,000 screening examinations in the included analyses.
The estimated additional detection rates were approximately 1.69 per 1,000 for DBT, 2.30 for automated breast ultrasound, and 2.57 for handheld ultrasound. Evidence for contrast-enhanced mammography was more limited, although one study reported additional cancer detection comparable to MRI.
These findings support the potential value of supplemental imaging in selected populations, particularly when mammography sensitivity is reduced by dense tissue.
However, the studies varied in patient populations, designs, and screening approaches. The pooled estimates should not be interpreted as direct head-to-head comparisons of all modalities under identical conditions.
Additional cancer detection also does not automatically demonstrate reduced breast cancer mortality for every supplemental imaging strategy. The balance between increased detection, false-positive findings, costs, availability, and patient preferences remains important.
The professional and regulatory context should also be considered.
In the United States, amendments to the Mammography Quality Standards Act regulations have required breast-density notification since September 10, 2024. These requirements provide an example of regulatory attention to communicating mammography limitations, but they are United States legal requirements and do not automatically apply in Viet Nam.
Viet Nam's Ministry of Health guidance under Decision No. 3898/QD-BYT specifically recognizes reduced mammographic detection capability in women with dense breasts, particularly BI-RADS density categories C and D.
The guidance recommends combining breast ultrasound with mammography in the relevant screening circumstances because ultrasound may identify abnormalities not visible on mammography.
Healthcare organizations should apply this guidance within the appropriate national screening pathway, taking account of the patient's clinical circumstances, authorized services, and applicable professional requirements.
The objective is to ensure that breast density is recognized and managed appropriately without causing unnecessary alarm or assuming that every woman with dense breasts requires the same additional examination regardless of clinical context.
Measuring Cancer Detection Rather Than Counting Mammograms
A mammography service may perform thousands of examinations each year, but high procedure volume does not establish how reliably cancers are detected.
The number of completed mammograms describes service activity, not diagnostic sensitivity. Similarly, reporting more abnormal findings does not necessarily demonstrate better performance. A service may identify additional cancers while also increasing unnecessary recalls and investigations.
Performance auditing provides a more meaningful approach by connecting imaging interpretations with subsequent diagnostic outcomes.
The BI-RADS framework supports standardized collection and analysis of this information. Relevant indicators include cancer detection rate, recall or abnormal interpretation rate, positive predictive value, sensitivity, and specificity.
These indicators address different aspects of performance and should not be interpreted independently.
A relatively high recall rate may reflect appropriate caution in a particular patient population, but it may also increase unnecessary examinations and anxiety.
Conversely, an unusually low recall rate is not necessarily evidence of superior performance. It may reflect a lower-risk population or, in some circumstances, inadequate recognition of suspicious findings.
The purpose of auditing is to understand the balance between identifying clinically important cancers and minimizing unnecessary investigations. This requires reliable outcome data, appropriate definitions, and careful consideration of the population being examined.
Screening and Diagnostic Mammography Require Separate Audits
One of the most important safeguards in performance measurement is distinguishing screening from diagnostic mammography.
Screening generally evaluates people without symptoms, many of whom have a relatively low probability of breast cancer at the time of examination.
Diagnostic mammography investigates a known or suspected concern, including clinical findings or abnormalities identified during previous imaging.
The probability of cancer is therefore different between these populations. A diagnostic service assessing patients with suspicious findings may identify a higher proportion of cancers than a routine screening program.
This does not automatically mean that its radiologists are more accurate. The difference may reflect the characteristics of the patients referred.
Performance benchmarks developed for screening mammography should therefore not be applied directly to diagnostic examinations without considering the relevant definitions and populations.
The Breast Cancer Surveillance Consortium (BCSC) provides extensive benchmarking data for screening and diagnostic mammography.
Its screening performance tables, updated in April 2026, include analyses based on 3,693,799 screening examinations performed between 2011 and 2021.
These are historical United States data made available through updated statistical tables in 2026. They should not be described as measurements of Vietnamese mammography performance in 2026.
The BCSC data provide useful examples of established performance measures and the importance of appropriate comparison groups.
However, their interpretation in Viet Nam requires attention to differences in patient characteristics, screening organization, breast density, technology, disease prevalence, and access to cancer outcome information.
International benchmarks can provide context, but they should not be used as automatic judgments of local professional performance.
Cancer Detection Rate, Recall Rate and Positive Predictive Value
Cancer detection rate describes the number of cancers identified through a defined screening process relative to the number of eligible screening examinations. It is commonly expressed as cancers detected per 1,000 examinations.
A reliable calculation requires consistent definitions of cancers attributed to the screening examination and appropriate verification of subsequent diagnostic outcomes.
A higher cancer detection rate may reflect effective screening. However, it may also occur because the population has a higher prevalence of breast cancer or includes more individuals with elevated risk. The indicator should therefore be interpreted alongside patient characteristics and screening history.
Recall rate measures the proportion of screening examinations interpreted as requiring additional evaluation according to the selected audit definition.
A high recall rate may increase patient anxiety, healthcare costs, and additional investigations. Nevertheless, reducing recalls should not become an isolated performance objective.
If radiologists become reluctant to recommend further assessment because of pressure to achieve a low recall rate, clinically important abnormalities may be overlooked.
The relationship between recall rate and cancer detection is more informative than either indicator alone.
Positive predictive value (PPV) provides another perspective by examining how often a positive interpretation or recommendation is associated with confirmed cancer.
Mammography auditing, however, uses different PPV definitions.
PPV1 assesses cancer yield among screening examinations meeting the applicable definition of a positive initial assessment. PPV2 concerns examinations for which biopsy or tissue diagnosis is recommended. PPV3 concerns the proportion of biopsies actually performed that result in a cancer diagnosis.
These measures are related but not interchangeable. A service may appear to have different performance depending on which denominator is used.
For this reason, an audit should not present a single percentage labeled "positive predictive value" without explaining how it was calculated.
The definitions, follow-up periods, and inclusion criteria established by the selected BI-RADS or other appropriate audit methodology should be applied consistently.
When an organization changes reporting software, adopts updated BI-RADS terminology, or modifies outcome tracking, it should examine whether those changes affect the calculation of performance indicators.
An apparent improvement or deterioration may reflect a change in measurement methodology rather than a genuine change in diagnostic performance.
Sensitivity and Interval Cancers Depend on Reliable Follow-Up Data
Sensitivity describes the proportion of cancers in a defined evaluation population that are identified by the examination, using an appropriate reference standard and follow-up period.
Accurately estimating sensitivity requires information about both detected cancers and cancers diagnosed despite an initially negative examination.
This can be challenging in routine healthcare settings. A mammography facility may know how many patients received a cancer diagnosis after its abnormal examinations but lack information about patients who were subsequently diagnosed at another hospital.
Without sufficiently complete outcome ascertainment, calculated sensitivity may be misleading.
Interval cancers are particularly relevant to screening quality. These are cancers diagnosed after a screening examination that did not identify cancer and before the next scheduled screening examination, according to the definition and screening interval used.
Reviewing interval cancers can provide valuable information about possible limitations in image acquisition, interpretation, breast density, and tumor biology.
However, the diagnosis of an interval cancer does not automatically establish that the previous mammogram was interpreted incorrectly.
Some tumors may have been genuinely occult on earlier images. Others may have developed or progressed rapidly after the examination. A structured review should distinguish these possibilities where the available evidence permits.
Reviewers should also recognize the potential for hindsight bias. Knowing the location of a subsequently diagnosed cancer can make a subtle earlier finding seem more obvious than it would have been during the original interpretation.
This is important when evaluating suspected missed cancers or drawing conclusions about individual professional performance.
For healthcare organizations in Viet Nam, more reliable sensitivity and interval cancer auditing may require better connections between imaging records, pathology results, clinical follow-up, and cancer outcome information.
Where follow-up data remain incomplete, the limitations of the audit should be stated rather than presenting uncertain estimates as definitive.
Breast Density and Case Mix Matter in Performance Comparisons
Mammography performance varies across patient populations.
Breast density is one contributing factor, but it is not the only one. Age, individual breast cancer risk, previous breast cancer history, screening history, and the distinction between initial and subsequent screening rounds may influence the frequency and characteristics of detected abnormalities.
Imaging technology also matters. Performance data obtained using conventional digital mammography should not automatically be compared with digital breast tomosynthesis without considering differences in acquisition, interpretation, and patient selection.
Where sample size and data quality permit, organizations may examine performance within relevant subgroups.
For example, comparing results among women with dense and nondense breasts may help identify particular detection challenges. Separating initial screening examinations from subsequent screening rounds may also provide more meaningful information than combining them into a single performance estimate.
However, subgroup analysis should not create incentives to avoid patients who are more difficult to image or who have complex clinical needs.
A service caring for patients with implants, previous surgery, mobility limitations, or extremely dense breasts may encounter more technically challenging examinations. Quality improvement should address those challenges rather than penalize professionals or facilities without accounting for the circumstances.
Small sample sizes require caution as well. An individual radiologist's cancer detection rate may vary substantially when relatively few examinations or cancers are included.
Meaningful comparisons should therefore consider case volume, case mix, statistical uncertainty, and the completeness of outcome information.
Connecting Technical Quality With Clinical Outcomes
Technical image-quality audits and cancer detection audits provide complementary information.
One evaluates whether mammographic images meet appropriate acquisition and anatomical coverage criteria. The other examines how effectively the service identifies cancer in a defined patient population.
A technically satisfactory mammogram does not guarantee that a cancer will be detected. Conversely, a department may detect a substantial number of cancers while still having recurring positioning or image-quality deficiencies requiring improvement.
The two forms of evidence should therefore be reviewed together.
Consider a service where positioning audits repeatedly identify inadequate inclusion of posterior breast tissue on MLO views. The organization should investigate this pattern even if its overall cancer detection rate appears acceptable.
Potential contributing factors may include practical technique, staff training, equipment configuration, scheduling pressures, or inconsistent criteria for deciding whether an image should be repeated.
Another service may consistently produce technically adequate images but experience unexpectedly low detection performance. In that situation, the organization may need to examine interpretation practices, access to previous images, reader workload, peer review, or the completeness of cancer outcome data.
The response should follow the evidence.
A low detection indicator does not automatically justify requiring every radiologist to undertake additional training. The underlying issue may involve incomplete follow-up data or differences in the patient population.
Similarly, introducing a new mammography system will not necessarily resolve positioning problems arising from workflow or practical technique.
Meaningful improvement depends on identifying which part of the diagnostic process is contributing to the observed limitation.
Professional Review, Competence and Organizational Accountability
Mammography quality depends on coordination among radiologic technologists, radiologists, medical physicists, biomedical engineering personnel, and clinical leadership.
Each group contributes to different aspects of the examination.
Technologists obtain technically appropriate images within their authorized professional scope. Radiologists interpret the images, characterize findings, and assess relevant diagnostic limitations.
Medical physicists and technical specialists contribute to equipment quality assurance, radiation dose assessment, and radiation protection. Clinical leadership and quality management personnel support competence assessment, performance review, incident investigation, and improvement activities.
These responsibilities need to be coordinated.
A radiologist who repeatedly identifies inadequate positioning should have an effective way to provide feedback to the technologists involved. Similarly, technologists should be able to raise concerns about equipment performance or difficult positioning circumstances without being discouraged from reporting them.
A review process focused only on individual mistakes may overlook wider system problems. Recurring positioning deficiencies may be associated with staff workload, inadequate appointment time, insufficient practical coaching, or equipment that is poorly suited to particular patient needs.
Professional peer review can also help radiologists learn from interpretive discrepancies.
However, disagreement between radiologists does not automatically establish that one interpretation was wrong.
Review should use appropriate case selection, clear assessment criteria, and methods for evaluating clinically meaningful differences. When a missed or potentially missed cancer is suspected, the assessment should consider the technical quality of the original examination and the information reasonably available at that time.
The objective is to identify opportunities for learning while avoiding unsupported judgments about individual professional performance.
Evidence involving Vietnamese clinicians also offers relevant insight.
A study by Trieu and colleagues, published online in October 2025 and in the 2026 volume of the Asia-Pacific Journal of Clinical Oncology, evaluated a digital breast tomosynthesis training workshop involving ten pairs of Vietnamese readers.
The investigators observed significant improvements in selected image-reading performance measures following structured educational activities. The study included radiologists and other physicians involved in breast imaging interpretation.
Although the findings suggest that targeted training can help strengthen interpretive skills, the evaluation involved controlled image-reading exercises. It does not independently demonstrate reductions in missed cancers or improvements in patient outcomes during routine clinical care.
The study nevertheless supports the value of considering local training needs and professional development when designing mammography quality improvement programs.
International quality frameworks can also support organizational improvement.
The Quality Standard for Imaging (QSI) 2026, developed jointly by the Royal College of Radiologists and the College of Radiographers in the United Kingdom, provides an updated framework for imaging governance, patient-centered care, safety, and continuous improvement.
It can offer useful guidance for imaging services in Viet Nam, but it should not be confused with a mandatory Vietnamese accreditation or licensing requirement.
Applying International Evidence Within Viet Nam's Healthcare System
The professional and regulatory context for mammography in Viet Nam deserves particular attention.
On December 18, 2025, the Ministry of Health issued Decision No. 3898/QD-BYT, promulgating the professional document Guidance on Breast Cancer Screening and Early Detection.
The guidance addresses breast cancer risk assessment, screening methods, mammography, breast ultrasound, magnetic resonance imaging, BI-RADS classification, and further diagnostic evaluation.
For mammography, it describes standard CC and MLO projections, breast compression, the importance of appropriately trained personnel, and limitations associated with dense breast tissue and other circumstances.
It also recognizes the reduced sensitivity of mammography in women with BI-RADS breast-density categories C and D and recommends combining ultrasound with mammography in the relevant screening pathway.
This provides an important national professional reference for healthcare organizations offering breast imaging services.
At the same time, the Ministry of Health guidance and ACR BI-RADS v2025 have different purposes.
The Vietnamese document provides national professional guidance concerning screening and early detection. BI-RADS provides detailed standardized terminology, assessment categories, reporting structures, and auditing methods.
Healthcare organizations may use BI-RADS to strengthen reporting consistency and outcome measurement, but should not assume that every United States audit benchmark or regulatory requirement automatically applies in Viet Nam.
Implementation should reflect the facility's authorized activities, clinical circumstances, equipment, professional competencies, and applicable Vietnamese requirements.
Screening practices must also be distinguished from diagnostic assessment of symptomatic patients.
A patient with a concerning breast symptom may require appropriate evaluation even when a recent screening mammogram was reported as negative.
A reassuring screening result should not be used to dismiss a persistent or clinically significant concern without suitable professional assessment.
Healthcare Licensing and Radiation Protection
Mammography involves ionizing radiation and is subject to professional, technical, and radiation safety requirements.
The broader healthcare framework includes the Law on Medical Examination and Treatment No. 15/2023/QH15 and Decree No. 96/2023/ND-CP, as amended where applicable.
These instruments are relevant to healthcare facility licensing, authorized professional activities, and conditions for providing clinical services.
Circular No. 23/2024/TT-BYT establishes the list of medical examination and treatment techniques and was amended by Circular No. 25/2026/TT-BYT.
Although Circular No. 25/2026/TT-BYT generally took effect on August 15, 2026, its Article 3 amendments concerning the medical examination and treatment technique lists took effect earlier, on July 1, 2026.
The amended provisions are reflected in Consolidated Document No. 13/2026/VBHN-TT-BYT dated July 30, 2026.
Under the amended framework, the list of techniques in Appendix 01 continues to apply through December 31, 2027, while the revised list in Appendix 02 is scheduled to apply from January 1, 2028.
Healthcare organizations should assess how these provisions and any relevant transitional arrangements apply to their authorized services.
Radiation protection is governed by a separate but related legal framework.
Viet Nam's Law on Atomic Energy No. 94/2025/QH15 took effect on January 1, 2026. Decree No. 332/2025/ND-CP, also effective January 1, 2026, provides implementing provisions concerning radiation safety, authorization, inspection, and related responsibilities.
Healthcare organizations should determine the specific requirements applicable to their mammography equipment, radiation-related activities, and operating circumstances.
These may include requirements concerning equipment operation, maintenance, quality assurance, radiation protection, inspection, and appropriately authorized personnel.
International technical standards and professional recommendations can support these activities, but they do not replace applicable Vietnamese legal requirements.
Medical Records, Outcome Tracking and Information Governance
Digital mammography involves the collection, storage, interpretation, and transfer of clinical images and related patient information.
Reliable information management supports both patient care and quality measurement.
A mammography service may require access to previous images, subsequent ultrasound or MRI findings, pathology reports, and information about cancers diagnosed after screening.
These connections allow radiologists and quality teams to understand how findings evolve over time and whether the original examination addressed its intended purpose. They are also important for calculating cancer detection performance accurately.
However, information must be handled according to applicable legal and organizational requirements.
Viet Nam's Law on Personal Data Protection No. 91/2025/QH15 and Decree No. 356/2025/ND-CP took effect on January 1, 2026.
These instruments are relevant to lawful personal data processing, confidentiality, security, authorized access, disclosure, and data transfers.
Healthcare organizations should establish suitable safeguards for mammography images, reports, and related clinical records.
Where information is used for quality audits, professional review, education, or research, the organization should determine the applicable legal basis and requirements for that activity.
The fact that a clinical image is useful for quality improvement does not automatically authorize unrestricted access, external sharing, or reuse for unrelated purposes.
Reliable information governance should support legitimate professional review while protecting patient confidentiality.
These arrangements are particularly important when facilities exchange images, refer patients for further assessment, or attempt to connect imaging findings with pathology and cancer outcome databases.
Building a Practical Mammography Quality Improvement Program
A healthcare organization does not necessarily need an advanced analytics platform to begin improving mammography reliability.
A useful starting point is a structured review of a representative sample of examinations.
The organization can assess whether images demonstrate adequate anatomical coverage, whether relevant artifacts are present, and whether technical limitations were recognized and addressed.
Assessment criteria should be sufficiently clear for different reviewers to apply consistently.
Findings can then be examined by projection, equipment, relevant patient characteristics, or other factors that may influence image quality.
When recurring problems are identified, the organization should investigate their likely causes.
A pattern of inadequate posterior tissue inclusion may suggest a need for supervised practical coaching, clearer positioning criteria, or adjustments to workflow. Frequent motion artifacts may require attention to patient communication, positioning, compression, or the circumstances of image acquisition.
Equipment-related concerns may require assessment by qualified technical personnel.
Improvement interventions should be selected according to the evidence rather than assuming that every problem can be resolved through additional training.
Subsequent review can determine whether image quality has improved and whether changes have introduced new difficulties.
The program should also consider the consistency of breast-density assessment and reporting.
When reliable clinical outcome information is available, performance audits can be expanded to include cancer detection rate, recall rate, and appropriately defined positive predictive values.
More demanding measures, including sensitivity and interval cancer rates, should be developed only when sufficiently complete follow-up information can be obtained.
It is preferable to report a limited number of dependable indicators than a larger collection of percentages whose denominators or clinical significance remain uncertain.
As data systems improve, the organization can examine the relationship between technical image quality, breast density, imaging interpretation, and verified clinical outcomes.
Feedback from technologists and radiologists should also form part of the improvement process because practical difficulties may not be apparent from numerical indicators alone.
An effective quality program connects identified problems with professional review, appropriate corrective action, and subsequent reassessment.
The objective is continuous learning and more dependable diagnostic care, not simply demonstrating compliance with a technical checklist.
Strengthening Mammography Reliability as Part of Diagnostic Excellence
Mammography can contribute substantially to breast cancer detection, but the reliability of an individual examination cannot be inferred from its completion alone.
Correct positioning helps ensure that relevant breast tissue is included in the images. Appropriate technical quality supports the recognition of abnormalities. Understanding breast density helps clinicians recognize when cancers may be obscured despite technically satisfactory imaging.
Performance auditing adds another essential dimension by examining whether imaging services identify cancers appropriately within defined populations and whether technical or interpretive practices require improvement.
The evidence available in 2026 provides useful foundations for this work.
ACR BI-RADS v2025 strengthens standardized breast imaging interpretation and outcome monitoring. Recent research offers more consistent approaches to image-quality assessment, while BCSC data provide established methods for evaluating screening performance.
Studies of supplemental imaging help clarify the limitations associated with dense breast tissue, and research involving Vietnamese clinicians highlights the potential value of targeted professional training.
In Viet Nam, Decision No. 3898/QD-BYT provides an important national professional framework for breast cancer screening and early detection. Applicable healthcare and radiation safety legislation establishes further responsibilities that facilities must assess according to their operating circumstances.
The objective should not be to maximize the number of mammograms performed, minimize recalls at any cost, or assume that the newest imaging technology always provides the best examination.
It should be to obtain images that adequately represent the relevant breast tissue, recognize the limitations of the method, interpret findings appropriately, and use dependable evidence to evaluate performance.
A normal mammogram cannot guarantee that breast cancer is absent. Dense tissue may reduce the visibility of disease, and an apparently favorable audit result may be misleading when patient characteristics or outcome data have not been adequately considered.
For healthcare organizations, diagnostic excellence involves recognizing these uncertainties and establishing practical systems to improve the quality and reliability of the information produced.
Reliable mammography depends on more than technically acceptable images. It requires adequate anatomical coverage, appropriate interpretation of breast density, and meaningful performance audits that help healthcare organizations strengthen cancer detection while minimizing avoidable diagnostic harm.
Selected References and Regulatory Sources
Newell MS, Destounis SV, Leung JWT, DeMartini WB, Lee CH, Eby PR. ACR BI-RADS v2025 Manual. Reston, VA: American College of Radiology; 2025.
Destounis SV, Friedewald SM, Grimm LJ, Poplack SP, Sung JS. Mammography. In: ACR BI-RADS v2025 Manual. Reston, VA: American College of Radiology; 2025.
Eby PR, Bennett DL, Cohen EO, Plecha DM, Sickles EA. Auditing and Outcomes Monitoring. In: ACR BI-RADS v2025 Manual. Reston, VA: American College of Radiology; 2025.
American College of Radiology. ACR Publishes BI-RADS v2025 Manual to Advance Breast Imaging Standards. December 1, 2025.
American College of Radiology. Mammography Positioning Improvement Collaborative and Mammography Positioning Review Guidelines. ACR Learning Network.
Vandendaele L, De Visschere PJ, Hendrickx S, et al. Image Quality Standards in Digital Mammography: A Consensus Statement. Radiography. 2026;32(4):103410. DOI: 10.1016/j.radi.2026.103410.
Singla V, Garg D, Sethi A. ACR BI-RADS v2025: A Summary of Changes. Current Problems in Diagnostic Radiology. 2026;55(6):836–856. DOI: 10.1067/j.cpradiol.2026.06.009.
Duggan SN, Azharuddin M, Hernández R, et al. Supplemental Imaging Modalities for Breast Cancer Screening in Women With Dense Breasts: A Systematic Review With Economic Considerations. The Breast. 2026;85:104668. DOI: 10.1016/j.breast.2025.104668.
Trieu PD, et al. Enhancing Diagnostic Precision in Breast Cancer Detection: A Digital Breast Tomosynthesis Workshop Approach for Clinicians in the Southeast Asian Population. Asia-Pacific Journal of Clinical Oncology. 2026;22(2):296–307. Published online October 4, 2025. DOI: 10.1111/ajco.70033.
Breast Cancer Surveillance Consortium. Screening Mammography Performance Benchmarks. Statistical tables updated April 22, 2026, based on screening examinations performed between 2011 and 2021.
Breast Cancer Surveillance Consortium. Diagnostic Mammography Performance Benchmarks.
Royal College of Radiologists and College of Radiographers. Quality Standard for Imaging: QSI 2026.
United States Food and Drug Administration. Final Rule to Amend the Mammography Quality Standards Act Regulations. Published March 10, 2023. Relevant compliance requirements, including breast-density notification, effective September 10, 2024.
Ministry of Health of Viet Nam. Decision No. 3898/QD-BYT, Promulgating Professional Guidance on Breast Cancer Screening and Early Detection. December 18, 2025.
National Assembly of Viet Nam. Law on Medical Examination and Treatment No. 15/2023/QH15. Effective January 1, 2024, as subsequently amended.
Government of Viet Nam. Decree No. 96/2023/ND-CP, detailing certain provisions of the Law on Medical Examination and Treatment, as amended where applicable.
Ministry of Health of Viet Nam. Circular No. 23/2024/TT-BYT, Promulgating the List of Medical Examination and Treatment Techniques. October 18, 2024, as subsequently amended.
Ministry of Health of Viet Nam. Circular No. 25/2026/TT-BYT, amending and supplementing provisions of several Ministry of Health circulars, including Circular No. 23/2024/TT-BYT. Issued June 30, 2026. Article 3 amendments effective July 1, 2026; general effective date August 15, 2026.
Ministry of Health of Viet Nam. Consolidated Document No. 13/2026/VBHN-TT-BYT, Concerning the List of Medical Examination and Treatment Techniques. July 30, 2026.
National Assembly of Viet Nam. Law on Atomic Energy No. 94/2025/QH15. Effective January 1, 2026.
Government of Viet Nam. Decree No. 332/2025/ND-CP, detailing certain provisions and measures for implementing the Law on Atomic Energy concerning radiation safety and related regulatory requirements. Effective January 1, 2026.
National Assembly of Viet Nam. Law on Personal Data Protection No. 91/2025/QH15. Effective January 1, 2026.
Government of Viet Nam. Decree No. 356/2025/ND-CP, detailing certain provisions and implementation measures of the Law on Personal Data Protection. Effective January 1, 2026.
Professional Scope and Disclaimer
This article is published by Diagnostic Excellence Vietnam (DxEx.vn), a professional initiative dedicated to advancing diagnostic quality, safety, and reliability in Viet Nam through evidence-informed knowledge sharing, risk management, measurement, and quality improvement.
The content is intended for general professional education and discussion. It does not constitute individualized medical advice, a breast cancer screening or diagnostic protocol, a radiation safety assessment, an accreditation evaluation, certification, a determination of professional or technical error, or legal advice.
The discussion primarily addresses mammography positioning, image quality, breast density, and detection performance auditing in screening and relevant diagnostic settings. Specialized circumstances involving breast implants, previous breast surgery, contrast-enhanced mammography, and other complex examinations may require additional professional guidance.
A technically adequate mammogram does not exclude every breast cancer. Breast-density classification, screening indicators, and imaging findings must be interpreted within the appropriate clinical context.
International standards, ACR BI-RADS guidance, professional quality frameworks, and performance benchmarks should be applied according to their defined purposes, populations, and limitations. They do not automatically establish statutory obligations or universal performance targets for all healthcare facilities in Viet Nam.
Vietnamese professional guidance, healthcare regulations, radiation safety requirements, and personal data protection obligations must be considered according to the organization's authorized activities and applicable legal circumstances.
Patient-specific decisions concerning mammography acquisition, image interpretation, breast-density assessment, supplemental examinations, diagnostic investigation, clinical follow-up, and treatment must be undertaken by appropriately qualified and authorized healthcare professionals in accordance with applicable Vietnamese law, relevant professional standards, validated institutional procedures, and the patient's clinical circumstances.
Diagnostic Excellence Vietnam does not provide mammography examinations, clinical diagnosis, or treatment services through this educational publication.
