AI Advances Breast Cancer Screening for Early Detection
Recent advancements in artificial intelligence (AI) are making significant strides in breast cancer screening, providing physicians with tools to identify women at a higher risk years before any signs of cancer emerge. Dr. Elisa Port, the Director of Breast Surgery at Mount Sinai and author of “The Breast Advice,” explains how AI is uncovering hidden tumors and reducing false positive results, particularly for women with dense breast tissue. This exciting development aims to encourage the 30-40% of women who currently avoid regular mammograms to participate in screenings.
A study conducted by researchers at NYU Langone Health and its Perlmutter Cancer Center introduced a new AI model that predicts a woman’s risk of developing breast cancer within the next five years. Unlike traditional methods that often rely on a single recent scan, this tool analyses multiple years of a woman’s 3D mammogram images to observe changes in breast tissue over time.
This innovative model was created using data from over 313,000 mammograms belonging to more than 161,000 women. It highlights the potential for AI to enhance the way doctors assess breast cancer risk. Connie Lehman, a breast imaging expert at Harvard Medical School who was not directly involved in the research, emphasized that AI could help identify women who might otherwise be overlooked in risk assessments.
“For many years, I’ve met women shocked to learn they have breast cancer despite no family history of the disease,” said Lehman. AI’s capability to analyze both genetic factors and intricate details within breast tissue might reveal risks that conventional evaluations miss.
The intuitive design of AI allows it to glean insights not just about the presence of cancer today but also about a woman’s future risk. This information could help doctors tailor screening plans to an individual’s risk level. For instance, those found to be at increased risk might be advised to undergo additional screenings like MRIs, while those at average risk could continue with standard mammograms.
One of the advantages of the AI tool is that it can re-evaluate mammograms that a woman has already completed, meaning patients wouldn’t need to undergo new scans or face additional radiation exposure.
While the AI model presents promising benefits, Lehman is quick to note that a higher risk score is not a definitive cancer diagnosis. Instead, it should encourage discussion with healthcare providers about further screening options. Conversely, a lower risk score does not negate the need for regular mammogram screenings.
Though the model shows promise, it remains a research tool and has not yet received FDA approval for clinical use. Researchers acknowledge that further testing is essential, as the current study utilized data from a specific set of hospitals and imaging equipment.
The team at NYU Langone is eager to broaden their research, planning to apply their AI model to data from various medical centers and diverse 3D mammography systems to validate its effectiveness across different populations.
