Artificial intelligence could reduce time to diagnose breast cancer

Share on facebook
Facebook
Share on google
Google+
Share on twitter
Twitter
Share on linkedin
LinkedIn

A new study is looking to reduce the length of time it takes to diagnose women who are at a higher risk of breast cancer using artificial intelligence.

Google Health has teamed up with Northwestern Medicine to explore whether artificial intelligence (AI) could prioritise reviews of mammograms with a higher suspicion of breast cancer.

The model is trained to quickly find mammograms which need further review and flags them to radiologists. Women whose mammograms show a higher likelihood of breast cancer might be able to be seen the same day for follow up, according to a statement from Northwestern Medicine.

Dr Sarah Friedewald, associate professor of radiology at Northwestern University Feinberg School of Medicine, said: “With the use of artificial intelligence, we hope to expedite the process to diagnosis of breast cancer by identifying suspicious findings on patients’ screening examinations earlier than the standard of care.

“Finding cancer earlier means it can be smaller and easier to treat. We hope this will ultimately save lives.”

The Goolge-funded study builds on research conducted by Northwestern Medicine, Google Health and the NHS in 2020, which found AI screening of mammograms was as accurate as human experts.

In some cases AI showed better accuracy than human experts, according to the research. Initial findings, published in Nature, suggest the AI can identify the disease with greater accuracy, fewer false positives and fewer false negatives.

Dr Mozziyar Etemadi, research assistant professor of anesthesiology at Northwestern Medicine, added: “This study is the next step by applying the AI models in a prospective study to better understand how AI can be the most helpful for clinicians and patients in the real world.”

Patients taking part in the study will continue to have their mammograms assessed by a radiologist, but the AI model will be applied to prioritise those who need follow ups.

If the tool determines a patient’s mammogram shows areas suspicious for cancer, radiologists will be able to conduct an immediate review of the images.

IT systems not up to scratch

A 2019 review into the UKs screening programmes found IT systems “cannot support the safe running of programmes” and need to be upgraded “urgently”.

Professor Sir Mike Richards found breast cancer screening programmes were often given “low priority” by NHS trusts and, in some cases, management systems have not been updated since the 2017 WannaCry cyber-attack.

But he said new technologies, like artificial intelligence (AI), had the potential to ease a growing pressure on the NHS workforce and are likely to bring benefit to healthcare in the coming years.

Currently each mammogram is independently evaluated by two radiographers, but AI has the potential to ease current workforce strains by taking on the job of one radiographer, he found.

The review was commissioned after 450,000 women aged between 68 and 71 were not invited to their final breast screening between 2009 and 2018 due to a failure to ensure IT systems could deliver a new algorithm. Up to 270 women died due to not receiving their screening invitation.


We are the hosts of Digital Health Rewired 2021, the must-attend virtual festival celebrating the best of digital health and care, taking place across 15-19 March 2021. Get a flavour of Rewired 2021 from the video below. You can book your place here

More to explorer