Skip to main content

Guidance on evaluating AI for use in breast screening

Posted by: and , Posted on: - Categories: General

As discussed in work to incorporate Artificial intelligence (AI) into NHS breast screening, AI-based technologies could potentially play a role in almost every stage of breast screening to:

  • improve quality of care by detecting cancers
  • help alleviate workforce pressures

However, if the wrong AI systems are adopted, they could increase the harms of breast screening, so it is important the guidance stays up to date with the technology.

Formal approach

The UK National Screening Committee (UK NSC) is the expert group which advises ministers in the 4 UK countries on all things screening. So, it is important that the UK NSC has the best possible AI guidance on which to properly assess any potential modifications to screening programmes.

The UK NSC has an AI task group which includes experts in AI, clinicians and public and patient members. This group has a vital role in horizon-scanning and looking at evidence for AI, concentrating on areas with most promise for early use in screening.

The AI task group developed the UK NSC Approach to Reviewing Evidence on AI in Breast Screening, which was recently published in the Lancet Digital Health, and has already received a lot of attention.

Potential impacts of AI

The new guidance communicates clearly to clinicians, policymakers, and AI developers what types of evidence is needed for the UK NSC to be sure that AI is safe and can be recommended for use.

The UK NSC needs to understand the potential impacts of AI because any significant change could alter the balance of benefit and harm of the breast screening programme. The new guidance provides a roadmap of how to measure whether AI will maximise benefit and minimise harm to women screened.

For example, what would the effects be if AI is shown to detect cancers which would never go on to cause women any harm? These effects could include unnecessary treatments and increased anxiety for women, potentially shifting the balance of harms and benefits.

Looking forward

A recent review examining the accuracy of AI systems for the detection of breast cancer in mammography screening concluded there is some way to go to achieve the quality and quantity required for the implementation of AI into clinical practice. But as the evidence starts to change, the UK NSC stands ready to react.

This fascinating area of research into AI in breast screening is one we’ll revisit on this blog, so do keep an eye out for updates.

UK NSC blog

The UK NSC blog provides up to date news from the UK National Screening Committee. You can register to receive updates direct to your inbox, so there’s no need to keep checking for new articles. If you have any questions about this blog article, or about the work of the UK NSC, please email

Sharing and comments

Share this page