AI Transforms Breast Cancer Screening, Aiding Early Detection

Artificial intelligence (AI) is revolutionizing the way breast cancer screenings are conducted, enabling doctors to identify women who may be at elevated risk of developing the disease years before it actually occurs. Dr. Elisa Port, who serves as the Breast Surgery Director at Mount Sinai and has authored ‘The Breast Advice,’ emphasizes that AI technology is improving the detection of hidden tumors while also reducing false positives—an aspect particularly beneficial for women with dense breast tissue. This breakthrough aims to encourage more women—currently around 30-40%—to participate in regular mammogram screenings.

Recent research published in the American Journal of Roentgenology highlights an AI model crafted by researchers at NYU Langone Health and its Perlmutter Cancer Center. This model is designed to estimate a woman’s risk of developing breast cancer within the span of five years.

Instead of focusing solely on the latest mammogram, the AI tool analyzes several years’ worth of a woman’s 3D mammogram data to observe changes in breast tissue over time, which is a novel approach.

The study utilized more than 313,000 mammograms from a diverse group of over 161,000 women. Connie Lehman, a specialist in breast imaging and a professor of radiology at Harvard Medical School, pointed out that this kind of AI could help detect women whose heightened risk might fly under the radar otherwise.

Lehman shared her insights, stating, “For decades, I have encountered women who were taken aback by a breast cancer diagnosis, often expressing disbelief with things like, ‘How can this be? No one in my family has had breast cancer.'” Traditional assessments often take into account factors like family history and genetic predispositions, but AI can dive deeper, analyzing the intricate details within breast tissue.

“A mammogram has the potential to offer more than just a snapshot of current health; it also holds vital clues regarding future risk,” she added.

This kind of information could eventually enable doctors to customize breast cancer screening protocols based on an individual’s specific risk profile. Women identified as being at higher risk can explore additional screening options, such as MRIs, with their healthcare providers, while those with average risk might continue with standard mammograms.

Interestingly, this technology could also reassess past mammograms, thus providing valuable risk information without necessitating additional imaging or radiation exposure.

However, Lehman did emphasize the limits of the AI risk assessment, making it clear that it is not an outright cancer diagnosis. “A higher score can ignite discussions about further screenings and risk management strategies,” she noted. “But a lower score doesn’t nullify the risk or replace the need for regular mammography.” The NYU model currently functions solely as a research tool and has not received FDA approval for clinical application yet.

The research team indicated that further testing is essential. The study utilized data only from NYU Langone hospitals and imaging equipment produced by a singular manufacturer. Future plans include validating the model across different medical facilities and utilizing various 3D mammography systems.

As the research progresses, it could lead to more personalized approaches to breast cancer screening, helping women better understand their health and make informed decisions. The endeavor stands as a testament to the potential of AI in shaping the future of medical diagnostics.