Potent logo

AI Model Enhances Precision in Breast Cancer Surgery, Predicts Tumour Removal Accurac

UNC Researchers Develop AI Tool to Aid Surgeons in Identifying Residual Cancer Cells

By Bianca LanePublished 2 months ago 4 min read

In the realm of healthcare, the marriage of artificial intelligence (AI) and machine learning has led to transformative breakthroughs that extend far beyond ethical discussions and appropriateness. One such groundbreaking collaboration between the UNC Department of Surgery, the Joint UNC-NCSU Department of Biomedical Engineering, and the UNC Lineberger Comprehensive Cancer Center has given birth to an extraordinary AI model. This model possesses the remarkable ability to predict whether cancerous tissue has been completely excised during breast cancer surgery, potentially obviating the need for follow-up surgeries and the associated physical and emotional toll on patients. The promising findings of this collaboration have been documented in the prestigious Annals of Surgical Oncology, marking a pivotal moment in the fight against breast cancer.

Breast cancer surgeries, which are a critical component of cancer treatment, typically involve the removal of tumors along with a portion of the surrounding healthy tissue to ensure the complete eradication of cancer. Post-surgery, specimens containing the excised tissue are meticulously examined to confirm the removal of the abnormal tissue. These specimens are then sent for pathological analysis, where highly trained pathologists determine if cancer cells extend to the specimen's outer edge, known as the pathological margin. If cancer cells are detected at this margin, it signals the potential presence of residual cancerous cells, necessitating additional surgery to remove the remaining malignancies.

The groundbreaking aspect of this collaborative effort lies in the AI model's remarkable ability to analyze specimens in real-time during surgery, eliminating the time lag traditionally associated with post-surgical pathological analysis. This innovative tool draws upon a vast database comprising hundreds of mammogram images of specimens, thoughtfully paired with final reports from pathologists. By examining this wealth of data, the AI model can effectively differentiate between specimens with positive and negative margins, providing surgeons with critical information during the surgery itself. In addition to image data, demographic information, including patient age, race, tumor type, and tumor size, has been integrated into the model to further refine its accuracy.

What sets this AI model apart is its exceptional performance, often rivalling or even surpassing human interpretations. This advanced technology excels in identifying positive margins, a crucial factor in avoiding follow-up surgeries. For patients, this represents a potential turning point in their cancer journey, as it significantly reduces the need for additional invasive procedures and the associated physical, emotional, and financial burdens.

Dr. Kevin Chen, the study's first author and a general surgery resident in the Department of Surgery, is particularly enthusiastic about the potential of AI to support surgeons' decision-making during procedures. He emphasizes that this innovation could prove especially valuable in cases where dense breast tissue obscures tumor margins on mammograms due to its bright white appearance, making it challenging to distinguish between cancerous and healthy tissue. In such scenarios, the AI model becomes an invaluable ally to the surgical team, offering real-time insights and enhancing the precision of surgical decisions.

Furthermore, this AI model has the potential to be a game-changer for hospitals with limited resources, serving as a valuable resource for surgeons, radiologists, and pathologists who may not have immediate access to specialized expertise. By harnessing the power of AI, hospitals can significantly enhance their decision-making processes during surgeries, ultimately contributing to more informed and precise outcomes for patients.

Dr. Shawn Gomez, co-senior author of the study and a professor of biomedical engineering and pharmacology, underscores the model's role as an additional layer of support within healthcare facilities. He highlights how it augments decision-making in hospitals that might lack immediate access to specialized knowledge or where the expertise of on-site pathologists is stretched thin. This AI tool equips surgeons with immediate feedback, thereby facilitating more informed and confident decisions during surgery, which is a pivotal aspect of achieving optimal outcomes.

While this AI model represents a remarkable leap forward in the field of breast cancer surgery, it is important to note that it is still in its early stages of development. Ongoing efforts include the continuous incorporation of additional patient images and input from various surgeons to enrich the model's capabilities. Furthermore, rigorous validation through extensive studies is essential before considering its deployment for clinical use. As researchers continue to expand their database and deepen their understanding of tissue appearances, tumor characteristics, and margins, the model's accuracy is expected to improve progressively over time.

In conclusion, the recent advancements in artificial intelligence and machine learning, as exemplified by the collaborative efforts of the UNC Department of Surgery, the Joint UNC-NCSU Department of Biomedical Engineering, and the UNC Lineberger Comprehensive Cancer Center, are poised to revolutionize breast cancer surgery. This innovative AI model, with its ability to provide real-time insights on tumor margins, holds the promise of reducing the need for follow-up surgeries and improving patient outcomes. While there is much work to be done to refine and validate this technology, it represents a significant step forward in the fight against breast cancer, offering hope and new possibilities for patients and healthcare providers alike.

fact or fictionhumanityhealthfeature

About the Creator

Bianca Lane

I'm passionate about writing diverse topics, like thrilling word adventures, where imagination knows no bounds. Exploring ideas and stories is my forte, and I'm eager to share them with you.

Reader insights

Be the first to share your insights about this piece.

How does it work?

Add your insights

Comments (1)

Sign in to comment
  • Alex H Mittelman 2 months ago

    Great work! Good job!

Find us on social media

Miscellaneous links

  • Explore
  • Contact
  • Privacy Policy
  • Terms of Use
  • Support

© 2023 Creatd, Inc. All Rights Reserved.