A new study has shown that using artificial intelligence (AI) in breast cancer screening can help doctors detect more cases of the disease. 

Advertisment

This is the first large-scale test of AI in a real-world screening programme, demonstrating that it can improve detection rates while not increasing false positives or unnecessary follow-up tests, as reported by The Guardian.

How the study was conducted 

The study, which analysed data from 461,818 women in Germany who underwent breast cancer screenings between July 2021 and February 2023, tested AI in a national screening programme targeting asymptomatic women aged 50 to 69. Each woman's mammogram was reviewed by two radiologists, with AI being used to assist in the review process for half of the women. The AI tool flagged suspicious areas and alerted radiologists if they missed something.

Advertisment

Also Read | Why is there no HMPV vaccine despite the virus's discovery 24 years ago?

AI boosts detection rates

According to the study published in the journal Nature Medicine, the results showed that women whose scans were assisted by AI had a 6.7 per cent higher breast cancer detection rate. After adjusting for factors like age and radiologist experience, the AI-assisted group detected 17.6 per cent more cancers. This means that one additional cancer case was found for every 1,000 women screened with AI.

Advertisment

Crucially, the AI tool did not lead to more false positives. This means women weren’t sent for unnecessary follow-up tests. The AI’s “safety net” function, which alerts radiologists when something suspicious is missed, led to 204 additional cancer diagnoses. Only 20 cancer cases would have been missed if radiologists had relied on their own assessments.

Also Read | Rabbit fever cases surge in the US: CDC report highlights risks

Helping radiologists work more efficiently  

The AI also helped reduce the workload of radiologists. Dr Stefan Bunk, co-author of the study and co-founder of Vara (the company behind the AI tool), explained that the AI allowed radiologists to review scans faster and focus on more complex cases. 

Expert reactions

Experts have had mixed reactions to the findings. Dr Stephen Duffy, emeritus professor of cancer screening at Queen Mary University of London, praised the results but noted that further research is needed to determine whether AI could safely replace the two-radiologist system currently used in the UK.

Dr Kristina Lång from Lund University raised concerns about overdiagnosis, particularly regarding slow-growing tumours that may not need treatment. She emphasised the importance of long-term studies to ensure AI is identifying clinically significant cancers.

(With inputs from agencies)