For the first time in history, an artificial intelligence system has guided surgeons through the removal of a brain tumour — helping doctors identify dangerous areas to avoid and, in doing so, preserving one patient's ability to see.
Rhys Hibbert, from Bedfordshire in England, became the world's first person to undergo this landmark procedure. Surgeons at London's National Hospital for Neurology and Neurosurgery used an AI system developed at University College London (UCL) to assist in removing as much of his brain tumour as possible while protecting the surrounding critical tissue.
The operation was carried out as part of a clinical trial. Hibbert said he was "humbled" to take part in the research. His surgeons said the outcome was the best they could have hoped for: the tumour was removed, and his eyesight was preserved.
"This is an example of AI at its best: patients getting care previously deemed unimaginable thanks to the latest groundbreaking technology," said James Frith, the UK's health innovation minister.
How the AI works
The system was trained on videos of previous brain operations. Over thousands of hours of surgical footage, it learned to recognise patterns in the tissue — what healthy brain looks like, where the margins of tumours tend to lie, and which structures need to be avoided.
During Hibbert's procedure, the AI analysed the surgical environment in real time, helping the team map the tumour's boundaries and identify structures near the visual pathway that, if damaged, could cause permanent sight loss. This was the key to saving his eye.
Critically, the AI did not perform the surgery. It acted as an expert guide — drawing on knowledge accumulated from far more operations than any individual surgeon could witness in a lifetime. The surgical team retained full control; the AI gave them better information to act on.
"It's a tool," one of the trial researchers explained, "like having a co-pilot who has studied thousands of journeys and can warn you of hazards ahead."
The challenge of brain tumour surgery
Removing brain tumours is among the most technically demanding procedures in medicine. The brain is a densely packed structure in which a few millimetres can separate a tumour from tissue responsible for speech, movement, memory, or vision. Surgeons must make real-time decisions under intense pressure, guided by pre-operative scans that can only tell part of the story.
Once inside the brain, tissue shifts, bleeds slightly, and changes in ways that static images cannot predict. Surgeons must rely on judgement, experience, and the live view through a microscope. Anything that improves the quality of information available in those moments can make the difference between a successful outcome and lasting damage.
The UCL system offers precisely that: a continuously updating analysis, trained on what actually happens in real operations, that highlights risks the surgeon might not see unaided.
What comes next
The clinical trial is continuing, and the data gathered from each operation — including Hibbert's — will help researchers refine the system further. The goal is to demonstrate that AI-guided surgery consistently improves outcomes before seeking regulatory approval for broader use.
If the evidence supports it, the technology could eventually be deployed across NHS hospitals and beyond — making the highest standard of neurosurgical guidance available to more patients, in more places.
The development is part of a broader transformation in medicine, as AI systems prove their value not just in analysis and diagnosis but in the operating room itself. Brain surgery is perhaps the highest-stakes environment where this shift is taking place. For patients like Rhys Hibbert, who left hospital with his sight intact and his tumour removed, the results speak for themselves.
