> ## Content Index
> Fetch the complete content index at: https://www.metatalks.ai/llms.txt
> Use this file to discover other available public pages before exploring further.

# Tumour removed in what surgeons call the first brain operation assisted live by AI
- URL: https://www.metatalks.ai/tumour-removed-in-what-surgeons-call-the-first-brain-operation-guided-live-by-ai/
- Published: 2026-08-31T13:37:00.000Z
- Updated: 2026-08-31T13:37:00.000Z
- Author: Al
- Tags: News, Health AI, #newswire

**As the surgeons worked, the software shaded the regions where nerves and vessels most probably lay and marked the least risky places to lift the growth out. The surgeons kept full control of the procedure.**

A pituitary tumour has been taken out of a patient in what his surgical team calls [the first brain operation](https://www.bbc.com/news/articles/cjwg5n7y68xo?ref=metatalks.ai) anywhere carried out with live artificial-intelligence assistance.

The output ran on a second monitor in theatre, following the instruments as they moved through the live video feed. It works much like the face-unlock software on a mobile phone, except that what it picks out are key anatomical structures, many of them out of sight.

The operation goes in through a nostril: a slim camera on a rod, an endoscope, is passed up to the base of the skull, where layers of bone and thick membrane still hide the growth. Nearby sit the optic nerves, which control vision, and the carotid arteries carrying blood up into the brain.

Anatomy varies from one person to the next, and in operations of this type part of the tumour is left behind in 25-50% of cases and a major blood vessel is damaged in 0.5-2%. Those are the risks the tool was built against.

Building it meant tracing by hand, across hundreds of recordings of earlier pituitary procedures, the outline of every vessel and nerve, so the software could learn to identify the structures that matter most and be checked on how well it did so. Years of laboratory testing, paid for by the National Institute for Health and Care Research and Google, came before this operation.

A typical surgeon in the UK does 10 to 20 of these operations a year. Prof Hani Marcus, one of the two neurosurgeons who carried out the procedure at the National Hospital for Neurology and Neurosurgery alongside Danyal Khan, told the BBC the software has learned from more operations than most surgeons will watch or do across an entire career, and that it can serve as a highly skilled extra set of eyes.

The patient, Rhys Hibbert, a father of two from Bedfordshire, had a benign 11mm growth on the pituitary, the marble-sized gland at the skull base. It was pushing against his optic nerves and narrowing his side vision, and his surgical team said he would have risked going blind had the procedure not gone ahead.

Offered the chance to go first, with an AI tool the team had built itself, he said yes at once. Across the eight weeks since, he has found his vision getting better every few days.

The research team stresses that the tool only assists. Further ahead, Marcus told the BBC, the ambition is a surgical equivalent of ChatGPT: a second specialist voice in the theatre that surgeons could consult for guidance, or set aside if they took a different view.

A larger trial is next.