Neurosurgeons in London have used artificial intelligence in real time during an operation to remove a pituitary tumour. We look at what happened, what we know so far, and the questions this raises for people affected by brain tumours.

Artificial intelligence is increasingly becoming part of conversations about healthcare and research. This week, a team of neurosurgeons in London took an important step in exploring how it could support surgeons during an operation.

At the National Hospital for Neurology and Neurosurgery, part of University College London Hospitals (UCLH), surgeons used an artificial intelligence system to analyse live images during an operation to remove a pituitary tumour.

The AI analysed the live video from the endoscope used during surgery, helping the surgical team identify important structures around the tumour, including nerves and blood vessels. The operation was part of a clinical trial using technology developed at University College London and funded by the National Institute for Health and Care Research and Google.

What did the AI actually do?

It is worth being clear about what is meant by “AI-assisted surgery”.

The AI did not perform the operation. It did not make decisions about the patient’s treatment. The surgeon remained in control throughout.

Instead, the technology provided an additional source of information to the surgical team. It analysed what the surgeon was seeing in real time and highlighted important anatomy that could help the team understand which areas to avoid while removing the tumour.

This is different from the idea of an autonomous robot carrying out surgery. The technology is being explored as a tool to support the surgeon, not replace them.

What is new about this?

AI has already been researched for use in brain tumour surgery. Previous work has explored how computer vision can recognise and classify anatomy and tissue during surgery.

What makes this development particularly interesting is that the AI was used to analyse the live surgical view in real time during an operation, rather than relying on scans taken before surgery.

For the surgical team, this could potentially provide useful information at a moment when precision is particularly important.

But this is early research

It is important not to get ahead of the evidence.

This was a pituitary tumour operation, using a technology specifically being explored for endoscopic pituitary surgery. We cannot assume that the same technology can currently be used in the same way for other types of brain tumour surgery.

And although the operation was successful and the patient’s vision improved, we cannot say from this one case that the AI itself caused that outcome or that AI-assisted surgery is now proven to be safer or more effective. Larger and further studies will be needed to understand what difference this technology could make for patients.

A note about pituitary tumours

You may have seen this story described in the news as brain tumour surgery.

The tumour involved in this operation was a pituitary tumour. Pituitary tumours develop in the pituitary gland, which sits at the base of the brain, but they are not technically classified as brain tumours.

The terminology is sometimes used differently in news reporting, which can understandably be confusing for people affected by a pituitary tumour or a brain tumour.

If you’re living with a pituitary tumour and need information or support, please contact the Pituitary Foundation.

What could this mean for people affected by brain tumours?

For people living with a brain tumour, developments in research can be exciting. They can also bring questions.

Will new technologies change the way people are treated? What should patients be told about technologies being used during their care? How much involvement should patients have in deciding how new technologies are developed and used?`

These questions matter because research is ultimately about people.

At brainstrust, we believe people affected by brain tumours should have a voice in the research and innovation that could shape their future.

Patient and public involvement and engagement (PPIE) gives people affected by brain tumours an opportunity to share their experiences, priorities and perspectives with researchers. Lived experience brings something important to research that technology and data cannot provide on their own.

As AI develops, understanding what matters to patients will be an important part of the conversation.

You can read more about the surgery and the research behind it on the UCL website.

What do you think?

If AI could be used as an additional tool to support your surgeon during treatment, how would you feel about it?

Would you want to know exactly how it worked? What its limitations were? What would reassure you, and what might concern you?

And as these technologies develop, where do you think the boundaries should be?

We want to hear from people affected by brain tumours and their families.

Join the conversation with us on our Facebook and Instagram pages.

Want to know more about AI?

AI is becoming increasingly part of the world around us, including healthcare. If you’re curious about what AI is, how it works and what it could mean for people affected by brain tumours, our Artificial Intelligence Know How resource can help you get to grips with the basics.