The use of artificial intelligence in breast cancer screening reduces subsequent cancer diagnosis rates by 12%, leading to improved early detection rates, according to the first trial of its kind.
Researchers said the study is the largest to date investigating the use of AI in cancer screening. The study involved 100,000 Swedish women who underwent mammography screening and were randomly assigned to either an AI-assisted exam or a standard exam performed by two radiologists between April 2021 and December 2022.
The AI system worked by analyzing mammograms by radiologists, assigning low-risk cases to one reading and high-risk cases to two readings, and also highlighting suspicious findings to support the radiologists.
According to a study published in The Lancet, AI-powered mammography screening reduced cancer diagnoses by 12% in the years after a breast cancer screening appointment. There were 1.55 cancers per 1,000 women in the AI support group, compared to 1.76 cancers per 1,000 women in the control group.
More than four out of five cancer cases (81%) in the AI-assisted mammography group were detected at the screening stage, compared with just under three-quarters (74%) in the control group. Additionally, the incidence of aggressive subtype cancers was reduced by almost one-third (27%) in the AI-assisted group compared to the control group.
Dr. Kristina Lang of Sweden’s Lund University and lead author of the study said AI-assisted mammography could help detect cancer early, but there are caveats.
“Wide deployment of AI-assisted mammography in breast cancer screening programs will not only help reduce workload pressure on radiologists, but could also help detect more cancers, including more aggressive subtypes, at an earlier stage,” said Ran.
“However, the introduction of AI in healthcare must be done carefully, using tested AI tools and with ongoing monitoring to ensure we have the right data on how AI will impact different local and national testing programs and how it will change over time.”
Breast cancer is the leading cause of death in women between the ages of 35 and 50, with more than 2 million women diagnosed with breast cancer each year worldwide.
While this study showed the clear benefits of AI for mammography exams, the researchers do not support replacing medical professionals with AI, as the screening exam still requires at least one human radiologist to perform the interpretation with AI assistance.
Dr Saumya Moorthy, senior strategic evidence manager at Cancer Research UK, said the study results were promising but cautioned.
“While using AI to help interpret mammograms is more efficient, there are concerns that it could lead to some cancers being missed. This study helps address concerns, but the results are from a single center, so further research is needed to know for sure whether this helps save lives,” Moorthy said.
She added: “With the number of people diagnosed with cancer expected to rise in the coming years, innovations like this will be essential to improving the NHS, but it is important that they are properly recognized to ensure people suffering from cancer are helped, not harmed.”
Simon Vincent, chief scientific officer at Breast Cancer Now, said: “This first trial highlights the huge potential of AI to support radiologists in breast cancer screening. Screening is an important tool for early detection and the earlier the disease is detected, the more likely it is that treatment will be successful.”
“This study shows that early diagnosis can improve and save lives, which is why the trial launched last year in the UK to explore the use of AI within the NHS breast cancer screening system will be important in determining the safest and most effective way to use these tools to catch more cancers earlier.”
