TL;DR

Surgeons at NHNN, the specialist neurology hospital within the UCLH trust, have removed a pituitary tumour with an AI system analysing the operation’s video feed as it happened. The tool was developed at UCL and used within a clinical trial. Healthcare Today reports it as the first time AI has assisted a neurosurgeon in real time, and the patient’s eyesight improved markedly afterwards.

What the AI did

Rather than relying on scans taken before the operation, this system interpreted the live camera feed, highlighting the key anatomy around the skull base so the team could decide more precisely where to cut and what to avoid. The pituitary gland sits among densely packed blood vessels and the nerves that control vision, which is what makes these operations delicate.

The patient, 48-year-old Rhys Hibbert from Bedfordshire, volunteered to go first. His tumour, about 11 millimetres across, was found by chance after he collapsed and had a seizure on a walk, and it had already cost him his peripheral vision. Untreated, it threatened his sight entirely. He could see clearly on waking, and a week later was walking unaided without vision aids.

Who is behind it

Hani Marcus, the consultant neurosurgeon at NHNN who led the operation and a professor at UCL, thanked trial participants and said the team had used “UK homegrown AI to help reduce risk during these delicate surgeries”. James Frith, the minister for health innovation, called it “an example of AI at its best”.

Why it matters for NHS AI

Recent NHS AI coverage has centred on administrative tools such as note-taking scribes, where the patient watchdog found AI scribes getting diagnoses wrong. This case sits at the other end of the risk scale, in the operating theatre, and it has been run the way high-stakes tools should be: inside a clinical trial, with the surgeon making the decisions and the AI flagging risk.

A single successful case is not evidence of wider benefit. Healthcare Today’s report gives no trial size, timetable or outcome measures, and those will decide whether the approach spreads.

Looking forward

The next markers are published trial results and whether the tool is extended to other procedures where critical anatomy is hard to see. For UK medtech, a homegrown system proven in a leading NHS neurosurgery unit is a strong reference case, if the trial data holds up.