Despite widespread use of mammography, breast cancer remains the most common cancer among women and a leading cause of cancer-related death globally, with an estimated 2.3 million new cases and 670,000 deaths reported in 2022 [[3]]. Challenges with early detection, particularly of aggressive tumors, are driving researchers to explore new technologies; a new artificial intelligence model offers a promising step toward more personalized and effective risk assessment and preventative care.
Worldwide, approximately 2.3 million cases of breast cancer are diagnosed annually, resulting in roughly 670,000 deaths among women. Despite the use of mammography screenings, breast cancer remains the leading cause of cancer-related deaths in women, according to Christiane Kuhl, director of the Department of Radiology at RWTH Aachen University Hospital.
Kuhl explained that mammography’s limitations in detecting all breast cancers, or detecting them early enough, contribute to this statistic. Specifically, rapidly growing and aggressive tumors are often less visible in mammograms, and these are the tumors most likely to prove fatal.
However, a new technological advancement offers the potential to improve early detection. Researchers have developed an artificial intelligence (AI) model that can analyze mammographic images and accurately classify an individual’s risk of developing breast cancer within the next five years. This is a significant step forward in personalized risk assessment, which could lead to more effective preventative care.
The study found that women identified as high-risk by the AI model were four times more likely to develop breast cancer compared to those assessed as having normal risk. “Specifically, women who were designated as high risk developed breast cancer four times more often than those whose AI score was low,” Kuhl stated.
“With this AI, we can predict with much greater accuracy whether a particular person will develop breast cancer in the next five years, based on mammograms that appear normal and show no signs of breast cancer,” Kuhl added.
Challenges in Detecting Risk with Mammograms
Table of Contents
Current guidelines recommend that women between the ages of 50 and 75 undergo a mammogram every two years for breast cancer screening. However, individual risk levels vary considerably.
Kuhl advocates for individualized breast cancer prevention, noting that mammography’s accuracy also differs significantly among women. Denser breast tissue is associated with a higher risk of the disease and reduces the effectiveness of mammography in predicting that risk – a fact many women are unaware of.
For women with extremely dense breast tissue, magnetic resonance imaging (MRI) is already recommended as a supplemental screening tool to reliably and early identify breast cancer.
To help identify which women would benefit from an early MRI, the Clairity Consortium – an international partnership of 46 research centers across the United States, Canada, South America, and Germany – developed the Clairity Breast AI system.
AI Enhances Precision in Early Detection
The AI model was trained using hundreds of thousands of mammograms from North America, South America, and Europe to determine breast cancer risk.
Unlike traditional risk models, the algorithm doesn’t require information about family history, genetics, or lifestyle factors. It calculates the probability of breast cancer solely from the mammogram and categorizes women into risk levels based on predefined thresholds.
The AI not only recognizes the amount of glandular tissue but also its texture – how the tissue is arranged – another parameter that influences breast cancer risk.
“Only around 10% of women have extremely dense glandular tissue. The majority of women who develop breast cancer and receive a late diagnosis have less dense tissue,” Kuhl explained. She emphasized that the key advancement is the AI’s ability to determine, within seconds, whether a woman needs an MRI for early detection.
Starting Prevention Earlier, But For Whom?
In most countries, breast cancer screening typically begins at age 50 because the risk increases significantly with age, and the benefit of widespread screening from that age is statistically proven.
While younger women are less frequently diagnosed, they are more likely to develop aggressive tumors when they do get sick. “In fact, young women would particularly benefit from early detection, provided that this detection works,” Kuhl said.
Targeted Risk Assessment
Christiane Kuhl believes that simply lowering the general age for screening is not an effective approach. “If we simply lower the age for women invited to mammographic screening, we are not changing anything about the fundamental problem.”
Instead, she proposes a two-step process: “First, mammography for early detection; then, an AI analysis to determine the risk of disease over the next five years.”
If the algorithm indicates a particularly high risk, an MRI should be offered. “In these women, mammography is no longer necessary,” Kuhl underscored.
(os/cp)
