Artificial intelligence (AI) seems to be everywhere. AI already helps us navigate traffic, conduct web searches, and shop online, and is rapidly establishing a foothold in healthcare. PINK Breast Center is proud to say we are already using AI to help our patients, with the first and only AI software for mammography the FDA has cleared.
Background
Breast cancer screening technologies have dramatically improved in recent years. One example is digital breast tomosynthesis (DBT) or 3D mammography, which PINK has offered for years. 3D mammography gives physicians many more images and better image quality than 2D mammography did. But these benefits have also presented new challenges, because there is so much more for the doctor to review. 2D systems typically produce four images per patient, whereas 3D mammography generates hundreds of images, or slices, of each breast. The larger number of images provides more information, but the quantity of subtle findings can be overwhelming. Studies have shown that this detail can increase reading time for radiologists, and can ultimately lead to reader fatigue.[i] The opportunity is for AI to review these images, to focus the doctor’s attention at areas of concern, and to increase the doctor’s confidence when an image is normal.
As PINK patients know, breast cancer is the second leading cause of death from cancer among women, affecting 1 in 8 women worldwide.[ii] We also know that early diagnosis generally leads to better patient outcomes.[iii],[iv],[v] PINK’s experience with AI confirms what new research suggests: AI can enhance doctors’ abilities to detect cancers earlier, when more treatment options may exist.
Read on to learn how PINK uses AI, and how AI might benefit you.
How it works
PINK was the first in Central and Northern New Jersey to implement iCAD’s ProFound AI™, the first and only artificial intelligence software for mammography cleared by the FDA. This is the only AI software clinically proven to enhance breast cancer screening, reduce false positives, and reduce unnecessary recalls.[vi] It’s not just PINK; leading healthcare facilities, hospitals and imaging centers worldwide now use ProFound AI every day to enhance breast cancer detection.
ProFound AI’s algorithm analyzes each 3D mammography image to pinpoint suspicious areas for PINK radiologists to examine more closely. AI also provides radiologists with analytics, which represent the algorithm’s confidence that a lesion is malignant. This helps PINK radiologists make more confident clinical decisions to benefit patients. Since ProFound AI uses the latest in deep-learning AI, the technology also allows for continuously improved performance via ongoing product updates.
According to a large reader study recently published in Radiology: Artificial Intelligence, ProFound AI improved cancer detection by 8 percent and reduced the rate of false positives and unnecessary patient recalls by 7.2 percent.1 Thus, AI not only enhances our ability to find cancers, it can reduce the false positives, which can be stressful for patients.
Where is it available?
Pink Breast Center is proud to offer 3D mammograms with ProFound AI in New Jersey, at PINK Flemington and PINK Paterson. If you or a loved one can’t make it to PINK, please visit this link to find a facility in your area that offers mammograms with ProFound AI.
Authored by Lisa M. Sheppard, MD
Dr. Lisa Sheppard is a Board-Certified General Radiologist with a Certificate of Additional Qualification (CAQ) in Neuroradiology and a special interest in women’s imaging and breast intervention. Dr. Sheppard completed a Radiology fellowship at the University of Pennsylvania, was Chief Radiology Resident at the Newark Beth Israel Medical Center, earned her medical degree (MD) at Jefferson Medical College, and graduated from St. Joseph’s University in Philadelphia. Dr. Sheppard has educated physicians at Michigan State University, Cooper University Hospital, and the Newark Beth Israel Medical Center.
[i] Hoffmeister, J. (2018). Artificial Intelligence for Digital Breast Tomosynthesis – Reader Study Results. [White paper]. Accessed via https://www.icadmed.com/assets/dmm253-reader-studies-results-rev-a.pdf
[ii] Bray, F. et al. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J. Clin. 68, 394–424 (2018).
[iii] Tabár, L. et al. Swedish two-county trial: impact of mammographic screening on breast cancer mortality during 3 decades. Radiology 260, 658–663 (2011).
[iv] The Canadian Task Force on Preventive Health Care. Recommendations on screening for breast cancer in average-risk women aged 40–74 years. CMAJ 183, 1991–2001 (2011).
[v] Marmot, M. G. et al. The benefits and harms of breast cancer screening: an independent
review. Br. J. Cancer 108, 2205–2240 (2013).
[vi] Conant, E. et al. (2019). Improving Accuracy and Efficiency with Concurrent Use of Artificial Intelligence for Digital Breast Tomosynthesis. Radiology: Artificial Intelligence. 1 (4). Accessed via https://pubs.rsna.org/doi/10.1148/ryai.2019180096