A recent study led by the University of Aberdeen has demonstrated that the use of artificial intelligence (AI) can enhance breast cancer detection rates by more than 10%. This innovation emerged from an initiative linked to NHS Grampian, focusing on how AI software could assist healthcare professionals during routine breast screenings for over 10,000 women.
The AI tool known as Mia, developed by the medical technology company Kheiron, has shown promising results. It identifies potentially concerning areas on mammogram scans that may be too subtle for the human eye to detect. The study, published in the Nature Cancer journal on March 15, 2024, reported an increase in detection rates by 10.4% while also alleviating the workload of medical staff and speeding up notification times for patients.
Real-World Impact of AI in Screening
Yvonne Cook, a participant in the AI research project, experienced firsthand the potential benefits of this technology. In a routine appointment in 2023, she opted in to the AI study and subsequently received a recall for further imaging after the AI flagged an abnormality. “I just felt incredibly lucky that I was part of the research programme and that it had been picked up at this early stage,” she stated.
At her follow-up appointment, a consultant confirmed the AI’s findings, revealing a small, Grade 2 tumor that would have likely gone unnoticed until it had grown larger. Cook was promptly placed on medication to manage the tumor’s growth, followed by surgery. “Had the AI not picked up the small tumor when it did, it could have resulted in much more invasive surgery and prolonged recovery,” she noted.
Cook’s case underscores the study’s conclusion that earlier detection leads to better treatment outcomes.
Expert Insights on AI’s Role in Healthcare
Professor Gerald Lip, clinical director for breast screening in north-east Scotland, emphasized the significance of these findings. He stated that AI has the potential to “effectively support” healthcare services by increasing cancer detection rates and reducing the burden on medical professionals.
“The bottom line here is without AI, doctors would not have caught these cancers as early,” he explained. As the integration of AI into clinical practice evolves, Lip noted that translating these findings into everyday healthcare will be one of the operational challenges faced in the coming decade.
The research team believes that the results will contribute to ongoing discussions about the role of AI in healthcare. The expansion of such studies across various sites in the UK aims to further explore the benefits of AI in breast screening, potentially changing the landscape of early cancer detection and patient care.
