AI may help detect aggressive breast cancers earlier, researchers say
Artificial intelligence is showing promise in the fight against aggressive breast cancer, potentially offering earlier detection and easing the burden on overworked medical professionals. A large-scale trial in Sweden has revealed significant benefits from incorporating AI into the standard mammography screening process.
The Swedish Trial and Its Findings
The study, published in The Lancet, analyzed data from nearly 100,000 women between the ages of 40 and 80 who participated in Sweden’s national breast screening programme between April 2021 and December 2022. Mammography is currently considered the most effective method for early breast cancer detection and reducing mortality rates, with European guidelines recommending independent review of scans by two radiologists.
Researchers have long been concerned about these “interval cancers,” as they tend to grow more rapidly, are often more advanced when discovered, and present greater treatment challenges. The Swedish trial demonstrated a 12 percent reduction in interval cancers over two years in the group utilizing AI-supported screening.
How AI Was Integrated
In the trial, participants were randomly assigned to either standard screening with two radiologists or an AI-assisted approach. With the AI support, the technology flagged suspicious areas on mammograms, which were then reviewed and finalized by a single radiologist. This process not only improved detection rates but also significantly reduced radiologist workload.
Jessie Gommers, a researcher from Radboud University Medical Centre, explained that the AI served as a support tool, “highlighting suspicious areas in a mammogram that helped radiologists during interpretation,” while emphasizing that “the final recall decision was still made by the radiologists.” Earlier findings indicated a 44 percent reduction in radiologist workload, a benefit particularly valuable in healthcare settings facing staffing shortages.
Looking Ahead
While AI-assisted mammography is already being implemented in parts of Sweden and Denmark, researchers caution that it is not yet ready to fully replace human oversight. Further evidence is needed to fully understand its long-term impact and ensure consistent performance. Independent experts, like Dr. Olga Oikonomidou from the University of Edinburgh’s Cancer Centre, emphasize that the focus should be on enhancing radiologist efficiency rather than outright replacement.
Dr. Oikonomidou also warned that AI systems vary considerably and require rigorous testing comparable to that of new medications, and are not necessarily a low-cost solution.
Globally, breast cancer remains a significant health concern, with an estimated 2.3 million diagnoses and 670,000 deaths in 2022. In Europe alone, the World Health Organization (WHO) recorded nearly 558,000 new cases that year. One in 20 women will be diagnosed with breast cancer during her lifetime.
Frequently Asked Questions
How does AI help with mammography?
The AI flags suspicious areas on mammograms, assisting radiologists in their interpretation and potentially leading to earlier detection of cancers.
Does AI replace radiologists?
No, the AI acts as a support tool. Radiologists still make the final decision regarding recalls and diagnoses.
What is an “interval cancer”?
An interval cancer is a cancer diagnosed between routine screening appointments. These cancers tend to grow faster and are more advanced at diagnosis.
As AI technology continues to evolve, it will be interesting to see how it is further integrated into breast cancer screening programs and what impact it will have on patient outcomes.