AnswersPublished19 sources
How Real-Time AI Assisted a World-First Brain Tumour Operation
In May 2026, 48 year old Rhys Hibbert became the first reported patient to undergo a successful brain tumour removal with live AI assistance. The UCL developed system analysed live endoscopic video and colour coded anatomy such as nerves and blood vessels, helping surgeons identify structures to protect in the confi...
Published byEdited with GPT-5.6 LunaImages generated with GPT Image 1.5
Research answer
Key takeaways
- In May 2026, 48 year old Rhys Hibbert became the first reported patient to undergo a successful brain tumour removal with live AI assistance.
- The UCL developed system analysed live endoscopic video and colour coded anatomy such as nerves and blood vessels, helping surgeons identify structures to protect in the confined skull base region.
- Hibbert reported restored vision and later returned to work, but the broader safety benefits of AI assisted surgery still require validation beyond this reported case.
AI Prompt
openai.comCreate a landscape editorial hero image for this Studio Global article: What was the world’s first real-time AI-guided brain-tumor removal procedure at London’s National Hospital for Neurology and Neurosurgery, w. Article summary: In May 2026, surgeons at London’s National Hospital for Neurology and Neurosurgery carried out what UCL Hospitals described as the world’s first successful brain-tumour removal using live, real-time AI assistance. The pa. Topic tags: general, general web. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake numbers, clic
In May 2026, surgeons at London’s National Hospital for Neurology and Neurosurgery performed what University College London Hospitals described as the world’s first successful brain-tumour removal using live, real-time artificial intelligence assistance. The patient was 48-year-old Rhys Hibbert, who had an 11 mm tumour on his pituitary gland that was affecting his sight. 1
2
5
The procedure is notable less because AI “performed” the operation than because it supplied surgeons with an additional layer of real-time visual information during a delicate endoscopic procedure. The surgical team retained responsibility for every decision.
What happened during the operation?
Surgeons used an endoscope to reach the tumour at the base of the skull. The AI system, developed through work associated with UCL’s Hawkes Institute, analysed the live camera feed and highlighted or colour-coded important structures visible in the surgical field. These included anatomy that surgeons needed to recognise and avoid, such as nerves and blood vessels. 5
6
The pituitary region is technically demanding because critical structures sit close together. The optic nerves and major blood vessels are near the gland, so accurately distinguishing tissue and anatomy during the operation is essential to reducing the risk of serious complications. UCL and UCLH have previously described the broader research effort as focused on identifying small tumours and critical structures around the pituitary gland. 19
Who was the patient, and what was removed?
The patient was Rhys Hibbert, a 48-year-old man from Bedfordshire and a customer-service manager. Reports said he faced worsening or imminent problems with his vision because of an 11 mm, non-cancerous tumour growing on the pituitary gland. 3
6
7
After the operation, Hibbert said he could see the room clearly when he woke. Reports also said that he was walking independently within a week, later returned to his job and described a much wider field of vision. 5
What did the AI actually do?
The system was designed to interpret surgical video while the operation was taking place. Its reported functions included recognising:
- Critical anatomy, including nerves and blood vessels
- Surgical instruments
- Tissue and instrument interactions
- Structures that could be colour-coded on the live display for the operating team
5
6
14
That makes the technology an intra-operative guidance tool rather than a robotic surgeon. It did not make independent surgical decisions, manipulate instruments or replace the neurosurgeons’ judgment. The team remained in control throughout the procedure. 5
The most useful comparison is an expert “second pair of eyes”: software that can continuously examine the camera feed and draw attention to anatomy that may be difficult to identify quickly in a confined and changing surgical field.
How long had the system been developed?
The supplied reporting describes the AI as the result of roughly a decade of UCL-led development aimed at interpreting complex endoscopic surgical video in real time. It was trained using surgical-video data so it could recognise relevant anatomy, instruments and their interactions during an operation. 2
The evidence supplied for this report does not establish precise details about the system’s hardware, including specific NVIDIA Clara IGX configurations, or provide a verified breakdown of funding from the UK National Institute for Health and Care Research and Google. Those details should not be treated as confirmed here.
Why the human-in-control model matters
Brain surgery is a high-risk setting for automation. An AI system may identify patterns quickly, but the operating team must interpret those signals in context, account for the patient’s anatomy and decide how to proceed. In this operation, AI was positioned as decision support—not as a replacement for clinical expertise or accountability. 5
6
That distinction is central to the significance of the procedure. The reported achievement shows that AI can analyse live operative imagery quickly enough to provide useful situational awareness during human surgery. It does not, by itself, demonstrate that AI-assisted surgery is safer than conventional surgery across patients or procedures.
What happens next?
The team’s longer-term vision is an advanced surgical assistant that can draw on expert knowledge during an operation—described in some reporting as analogous to a “ChatGPT for surgeons”—while leaving judgment and control with clinicians. 13
The supplied evidence does not provide confirmed details about the size, timetable or endpoints of a planned larger clinical trial. Further prospective research would be needed to determine how reliably the system performs across different surgeons, patients and procedures, and whether its assistance improves outcomes.
For now, the London operation is best understood as an early but important demonstration: real-time AI can help map delicate anatomy during a brain-tumour procedure, while the surgeons remain responsible for the operation. Hibbert’s reported recovery is encouraging, but the broader promise of AI-assisted surgery still depends on clinical validation beyond a single landmark case.
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
People also ask
What is the short answer to "How Real-Time AI Assisted a World-First Brain Tumour Operation"?
In May 2026, 48 year old Rhys Hibbert became the first reported patient to undergo a successful brain tumour removal with live AI assistance.
What are the key points to validate first?
In May 2026, 48 year old Rhys Hibbert became the first reported patient to undergo a successful brain tumour removal with live AI assistance. The UCL developed system analysed live endoscopic video and colour coded anatomy such as nerves and blood vessels, helping surgeons identify structures to protect in the confined skull base region.
What should I do next in practice?
Hibbert reported restored vision and later returned to work, but the broader safety benefits of AI assisted surgery still require validation beyond this reported case.