Doctors bested by AI in breast cancer detection

Google's research could support radiographers in diagnosing the condition in future

Researchers at Google Health have developed an artificial intelligence (AI) program that could lead to better detection and treatment of breast cancer.

Working with DeepMind, Cancer Research UK Imperial Centre, Northwestern University and Royal Surrey County Hospital, the organisation set out to create a piece of software that could spot breast cancer in mammograms more accurately than expert radiologists.

Advertisement - Article continues below

The AI was trained on a data set comprising mammograms from over 76,000 women in the UK and more than 15,000 women in the US. It was then tested on images from a different set of 25,000 UK and 3,000 US women.

Compared to radiologists, the AI produced a marked reduction in false positives for the US (5.7%), and a more modest reduction in false positives for the UK (1.2%). It was even more successful at reducing false negatives, producing a 9.7% fall in this type of error for the US and a 2.7% reduction in the UK.

The creators also ran a separate test to determine how effective it is across healthcare systems, by training it on only UK data then testing it against US data. In this scenario, it produced a 3.5% reduction in false positives and 8.1% reduction in false negatives.

The researchers believe this “sets the stage” for a future where radiologists are supported by AI in the diagnostic process for increased accuracy, particularly in the UK, which is facing a shortage of consultant radiologists in the NHS.

Advertisement
Advertisement - Article continues below
Advertisement - Article continues below

In a blog post, Shravya Shetty, technical lead of Google Health and Daniel Tse, product manager of Google Health, said: “There are some promising signs that the model could potentially increase the accuracy and efficiency of screening programs, as well as reduce wait times and stress for patients.”

They added, however, that “getting there will require continued research, prospective clinical studies and regulatory approval to understand and prove how software systems inspired by this research could improve patient care”.

The initial findings of the research have been published in full in the journal Nature.

Featured Resources

Top 5 challenges of migrating applications to the cloud

Explore how VMware Cloud on AWS helps to address common cloud migration challenges

Download now

3 reasons why now is the time to rethink your network

Changing requirements call for new solutions

Download now

All-flash buyer’s guide

Tips for evaluating Solid-State Arrays

Download now

Enabling enterprise machine and deep learning with intelligent storage

The power of AI can only be realised through efficient and performant delivery of data

Download now
Advertisement

Most Popular

Visit/security/privacy/355155/zoom-kills-facebook-integration-after-data-transfer-backlash
privacy

Zoom kills Facebook integration after data transfer backlash

30 Mar 2020
Visit/security/data-breaches/355173/marriott-hit-by-data-breach-exposing-personal-data-of-52-million
data breaches

Marriott data breach exposes personal data of 5.2 million guests

31 Mar 2020
Visit/security/cyber-crime/355171/fbi-warns-of-zoom-bombing-hackers-amidst-coronavirus-usage-spike
cyber crime

FBI warns of ‘Zoom-bombing’ hackers amid coronavirus usage spike

31 Mar 2020
Visit/data-insights/data-management/355170/oracle-cloud-courses-are-free-during-coronavirus-lockdown
data management

Oracle cloud courses are free during coronavirus lockdown

31 Mar 2020