Skip to Main Content
   
the NC AHEC logo    
     

Mammography Symposium 2025

September 17, 2025

Conference Brochure

                                                                       

AI in Breast Imaging

Artificial intelligence (AI) is rapidly transforming breast imaging, offering significant advancements across various modalities and stages of patient care. Its integration aims to enhance diagnostic accuracy, improve workflow efficiency, personalize patient management, and ultimately lead to better outcomes in breast cancer detection and treatment.

How AI is Transforming Breast Imaging

AI, particularly through machine learning and deep learning algorithms, analyzes complex patterns in large datasets of medical images. This capability allows AI to assist human readers and even perform certain tasks with high accuracy, often beyond the capabilities of the unaided human eye.

Here's a breakdown of AI's impact across different aspects of breast imaging:

Key Considerations and Future Directions:

While AI offers immense potential, ongoing research focuses on:

  • Validation: Rigorous validation of AI models in diverse, real-world clinical settings to ensure generalizability and prevent biases.

  • Explainable AI (XAI): Developing AI models that can explain their decisions to foster trust and facilitate human oversight.

  • Integration into Workflow: Seamless integration of AI tools into existing clinical workflows to maximize efficiency without disrupting practice.

  • Regulatory Approval: Navigating the regulatory landscape for AI-driven medical devices (like FDA approvals in the US).

The synergistic effect between human radiologists and AI systems holds the most promise, leading to improved patient care and potentially transforming the future of breast cancer management.

Source: Google Gemini

Research from PubMed:

Items 1-10 of 10 (Display the 10 citations in PubMed)

1.

Challenges in Implementing Artificial Intelligence in Breast Cancer Screening Programs: Systematic Review and Framework for Safe Adoption.

Goh S, Goh RSJ, Chong B, Ng QX, Koh GCH, Ngiam KY, Hartman M.

J Med Internet Res. 2025 May 15;27:e62941. doi: 10.2196/62941.

PMID: 40373301 Free PMC article. Review.

2.

Application of Artificial Intelligence in Cardio-Oncology Imaging for Cancer Therapy-Related Cardiovascular Toxicity: Systematic Review.

Mushcab H, Al Ramis M, AlRujaib A, Eskandarani R, Sunbul T, AlOtaibi A, Obaidan M, Al Harbi R, Aljabri D.

JMIR Cancer. 2025 May 9;11:e63964. doi: 10.2196/63964.

PMID: 40344203 Free PMC article. Review.

3.

Artificial Intelligence Performance in Image-Based Cancer Identification: Umbrella Review of Systematic Reviews.

Xu HL, Gong TT, Song XJ, Chen Q, Bao Q, Yao W, Xie MM, Li C, Grzegorzek M, Shi Y, Sun HZ, Li XH, Zhao YH, Gao S, Wu QJ.

J Med Internet Res. 2025 Apr 1;27:e53567. doi: 10.2196/53567.

PMID: 40167239 Free PMC article.

4.

Assessment of breast composition in MRI using artificial intelligence - A systematic review.

Murphy PC, McEntee M, Maher M, Ryan MF, Harman C, England A, Moore N.

Radiography (Lond). 2025 Mar;31(2):102900. doi: 10.1016/j.radi.2025.102900. Epub 2025 Feb 20.

PMID: 39983661

5.

Harnessing artificial intelligence for predicting breast cancer recurrence: a systematic review of clinical and imaging data.

Silveira JA, da Silva AR, de Lima MZT.

Discov Oncol. 2025 Feb 8;16(1):135. doi: 10.1007/s12672-025-01908-6.

PMID: 39921795 Free PMC article. Review.

6.

Deep learning-based breast cancer diagnosis in breast MRI: systematic review and meta-analysis.

Abdullah KA, Marziali S, Nanaa M, Escudero Sánchez L, Payne NR, Gilbert FJ.

Eur Radiol. 2025 Aug;35(8):4474-4489. doi: 10.1007/s00330-025-11406-6. Epub 2025 Feb 5.

PMID: 39907762 Free PMC article. Review.

7.

A Systematic Review of the Diagnostic Accuracy of Deep Learning Models for the Automatic Detection, Localization, and Characterization of Clinically Significant Prostate Cancer on Magnetic Resonance Imaging.

Molière S, Hamzaoui D, Ploussard G, Mathieu R, Fiard G, Baboudjian M, Granger B, Roupret M, Delingette H, Renard-Penna R.

Eur Urol Oncol. 2024 Nov 14:S2588-9311(24)00248-7. doi: 10.1016/j.euo.2024.11.001. Online ahead of print.

PMID: 39547898 Review.

8.

Advancing personalized oncology: a systematic review on the integration of artificial intelligence in monitoring neoadjuvant treatment for breast cancer patients.

Hachache R, Yahyaouy A, Riffi J, Tairi H, Abibou S, Adoui ME, Benjelloun M.

BMC Cancer. 2024 Oct 21;24(1):1300. doi: 10.1186/s12885-024-13049-0.

PMID: 39434042 Free PMC article.

9.

Artificial intelligence in mammography: a systematic review of the external validation.

Branco PESC, Franco AHS, de Oliveira AP, Carneiro IMC, de Carvalho LMC, de Souza JIN, Leandro DR, Cândido EB.

Rev Bras Ginecol Obstet. 2024 Sep 4;46:e-rbgo71. doi: 10.61622/rbgo/2024rbgo71. eCollection 2024.

PMID: 39380589 Free PMC article. Review.

10.

Frequency and characteristics of errors by artificial intelligence (AI) in reading screening mammography: a systematic review.

Zeng A, Houssami N, Noguchi N, Nickel B, Marinovich ML.

Breast Cancer Res Treat. 2024 Aug;207(1):1-13. doi: 10.1007/s10549-024-07353-3. Epub 2024 Jun 9.

PMID: 38853221 Free PMC article. Review.

Interpreting the breast pathology report

Presenting a mammography case effectively is crucial for clinical decision-making, peer review, and educational purposes. It requires a structured approach to convey all pertinent information concisely.

Here's a breakdown of case presentation fundamentals in mammography, with relevant links and URLs:

Case Presentation Fundamentals in Mammography

A strong mammography case presentation typically follows a logical flow, ensuring all critical information is covered for accurate assessment and recommendations.

  1. Patient Demographics and Clinical Context:

    • What to include: Patient's age, relevant medical history (e.g., personal or family history of breast cancer, prior biopsies, surgeries, hormone replacement therapy, breast symptoms).

    • Why it's important: Provides context for interpreting the images and assessing risk.

    • URL (General Patient History Relevance): While no single URL dictates "how to present patient demographics," the importance of clinical history is emphasized in practice guidelines.

  2. Reason for Examination:

    • What to include: Was it a screening mammogram (routine check-up for asymptomatic women) or a diagnostic mammogram (investigating symptoms like a palpable lump, pain, nipple discharge, or follow-up from an abnormal screening)?

    • Why it's important: Guides the extent of the examination and subsequent work-up.

  3. Prior Imaging Comparison:

    • What to include: Always compare with prior mammograms, and if available, prior ultrasound or MRI studies. Note the date of the most recent comparison.

    • Why it's important: Crucial for assessing stability, growth, or new findings. Changes are often the most significant indicator of pathology.

    • URL (Importance of Prior Imaging):

  4. Imaging Findings - Description:

    • What to include: Systematically describe any findings using standard mammographic terminology, often following the ACR BI-RADS lexicon.
      • Location: Quadrant (e.g., Upper Outer Quadrant - UOQ, Lower Inner Quadrant - LIQ), clock face position (e.g., 2 o'clock), and depth (anterior, middle, posterior).

      • Views: Specify the views on which the finding is seen (e.g., seen on MLO and CC views).

      • Masses: Shape (round, oval, irregular), Margin (circumscribed, microlobulated, obscured, indistinct, spiculated), Density (high, equal, low, fat-containing).
  1. Assessment (BI-RADS Category):

    • What to include: Assign the appropriate BI-RADS assessment category based on the likelihood of malignancy.

      • BI-RADS 0: Incomplete

      • BI-RADS 1: Negative

      • BI-RADS 2: Benign

      • BI-RADS 3: Probably Benign

      • BI-RADS 4: Suspicious (A, B, C subcategories)

      • BI-RADS 5: Highly Suggestive of Malignancy
      • BI-RADS 6: Known Biopsy-Proven Malignancy
    • Why it's important: Standardizes the interpretation and guides management recommendations.

    • URL (ACR BI-RADS Categories):

  2. Recommendations:

    • What to include: Clear recommendations for next steps, based on the BI-RADS assessment.

      • For BI-RADS 0: Recommend additional imaging (e.g., spot compression, magnification views, ultrasound) or comparison with prior studies.

      • For BI-RADS 3: Recommend short-interval follow-up (e.g., 6 months).

      • For BI-RADS 4/5: Recommend biopsy (e.g., core needle biopsy, excisional biopsy).

      • For BI-RADS 1/2: Recommend routine screening.

      • For BI-RADS 6: Recommend appropriate definitive treatment (e.g., surgical excision).

    • Why it's important: Provides actionable guidance for patient management.

    • URL (General Mammography Guidelines and Recommendations):

  3. Summary/Impression:

    • What to include: A concise summary of the most significant findings and the final BI-RADS assessment.

    • Why it's important: Offers a quick overview of the case for other clinicians.

Additional Resources for Mammography Professionals:

Source: Google Gemini

Case Presentation Fundamentals

  • What are the essential components of a mammography case presentation?
  • What are the best practices for structuring a mammography case presentation for peer review?
  • Mammography case presentation guidelines ACR
  • SBI mammography case presentation resources
  • BI-RADS lexicon official guide
  • Mammography reporting standards
  • Breast imaging clinical history importance

Presenting a mammography case effectively is crucial for clinical decision-making, peer review, and educational purposes. It requires a structured approach to convey all pertinent information concisely.

Case Presentation Fundamentals in Mammography

A strong mammography case presentation typically follows a logical flow, ensuring all critical information is covered for accurate assessment and recommendations, often adhering to standards like those set by the American College of Radiology (ACR) and utilizing the Breast Imaging-Reporting and Data System (BI-RADS) lexicon.

  1. Patient Demographics and Clinical Context:
  2. Reason for Examination:
    • What to include: Clearly state whether it's a screening mammogram (routine check-up for asymptomatic women) or a diagnostic mammogram (investigating symptoms, abnormal screening findings, or follow-up of a known lesion).
    • Why it's important: Dictates the appropriate imaging protocol and the urgency of the assessment.
  3. Prior Imaging Comparison:
    • What to include: Always compare with prior mammograms, and if available, prior ultrasound or MRI studies. Specify the dates and locations of these prior studies.
    • Why it's important: Crucial for assessing stability, growth, or the appearance of new findings, which are often the most significant indicators of malignancy.
    • URL (Importance of Prior Imaging):
  4. Imaging Findings - Description:
    • What to include: Systematically describe any findings using standard mammographic terminology from the ACR BI-RADS lexicon.
      • Breast Composition/Density: Report the BI-RADS density category (A: Almost entirely fatty, B: Scattered fibroglandular density, C: Heterogeneously dense, D: Extremely dense). This is a mandated reporting element.
      • Location: Specify the breast (right/left), quadrant (e.g., Upper Outer Quadrant - UOQ, Lower Inner Quadrant - LIQ), clock face position (e.g., 2 o'clock), and depth (anterior, middle, posterior).
      • Views: Indicate on which mammographic views the finding is seen (e.g., seen on MLO and CC views).
      • Masses: Describe Shape (round, oval, irregular), Margin (circumscribed, microlobulated, obscured, indistinct, spiculated), and Density (high, equal, low, fat-containing).
      • Calcifications: Describe Type (punctate, amorphous, pleomorphic, fine linear branching), and Distribution (diffuse, regional, clustered, segmental, linear).
      • Architectural Distortion: Description of any distortion without an associated mass.
      • Asymmetry: Description of focal, global, or developing asymmetry.
      • Associated Features: Skin retraction, nipple retraction, skin thickening, trabecular thickening, lymphadenopathy, edema.
    • Why it's important: Provides a clear, objective, and consistent description of abnormalities, ensuring reproducibility and understanding across different clinicians.
    • URL (ACR BI-RADS Atlas – Lexicon):
  5. Assessment (BI-RADS Category):
    • What to include: Assign the overall BI-RADS assessment category, which reflects the likelihood of malignancy based on all findings.
      • BI-RADS 0: Incomplete
      • BI-RADS 1: Negative1
      • BI-RADS 2: Benign
      • BI-RADS 3: Probably Benign
      • BI-RADS 4: Suspicious (subdivided into 4A, 4B, 4C for increasing suspicion)2
      • BI-RADS 5: Highly Suggestive of Malignancy3
      • BI-RADS 6: Known Biopsy-Proven Malignancy4
    • Why it's important: Standardizes the final interpretation and directly guides management recommendations.
    • URL (ACR BI-RADS Categories):
  6. Recommendations:
    • What to include: Clear, actionable recommendations for next steps, directly corresponding to the BI-RADS assessment.
      • For BI-RADS 0: Recommend additional imaging (e.g., spot compression, magnification views, ultrasound) or comparison with prior studies.5
      • For BI-RADS 3: Recommend short-interval follow-up (e.g., 6 months).
      • For BI-RADS 4/5: Recommend biopsy (e.g., core needle biopsy, excisional biopsy), specifying the imaging guidance.6
      • For BI-RADS 1/2: Recommend routine screening.
      • For BI-RADS 6: Recommend appropriate definitive treatment (e.g., surgical excision, as diagnosis is already confirmed).
    • Why it's important: Provides concrete guidance for the referring clinician and patient, ensuring timely and appropriate follow-up.
    • URL (General Mammography Guidelines and Recommendations):
  7. Summary/Impression:
    • What to include: A concise concluding statement that summarizes the most significant findings, the overall BI-RADS assessment, and the final recommendation.
    • Why it's important: Offers a quick, comprehensive overview of the case for other clinicians who may not read the full detailed report.

Additional Resources for Mammography Professionals:

  • Society of Breast Imaging (SBI): A leading resource for breast imaging professionals, offering educational content, guidelines, and case studies.7
  • American College of Radiology (ACR): Beyond BI-RADS, the ACR offers extensive resources for breast imaging quality and practice.8

Mammography and Breast Imaging Resources - American College of Radiology

Source: Google Gemini