First Major Trial Suggests AI Could Transform Breast Cancer Screening

email: online@newsofbahrain.com

Stockholm: The use of artificial intelligence in breast cancer screening has been linked to fewer cancer diagnoses in the years following screening and higher rates of early detection, according to the first large-scale trial of its kind.

The study, described by researchers as the largest to date examining AI in cancer screening, involved nearly 100,000 women in Sweden who participated in routine mammography screening between April 2021 and December 2022. Participants were randomly assigned to receive either AI-supported screening or standard screening read independently by two radiologists.

The AI system analysed mammograms to assess cancer risk, assigning low-risk cases to a single radiologist review and high-risk cases to double readings. It also highlighted suspicious areas to assist radiologists in identifying potential cancers.

According to findings published in The Lancet, women screened with AI support were 12 per cent less likely to be diagnosed with breast cancer in the years following a screening appointment. The rate of cancer diagnoses was 1.55 per 1,000 women in the AI-supported group, compared with 1.76 per 1,000 women in the standard screening group.

Early detection rates were also higher. More than four in five cancers (81 per cent) in the AI group were detected during screening, compared with 74 per cent in the control group. Researchers also reported nearly a 27 per cent reduction in aggressive breast cancer subtypes among women screened with AI.

Dr Kristina Lång of Lund University, the study’s lead author, said the findings suggest AI-supported mammography could improve early detection while easing pressure on radiologists.

“Wider adoption of AI-supported mammography could help reduce workload pressures and detect more cancers at an early stage, including aggressive subtypes,” she said. “However, introducing AI in healthcare must be done cautiously, using well-tested tools and continuous monitoring to understand how outcomes may vary across screening programmes over time.”

Breast cancer remains the leading cause of cancer-related death among women aged 35 to 50, with more than two million people diagnosed worldwide each year.

Despite the positive findings, researchers emphasised that AI should not replace healthcare professionals. Screening still requires at least one human radiologist to review mammograms alongside AI support.