A British father has become the world’s first patient to undergo brain surgery with live artificial intelligence assistance, with surgeons successfully removing his tumour using an AI system that analysed the operation in real time.
Rhys Hibbert, 48, from Bedfordshire, underwent the pioneering procedure at University College London Hospital after a non-cancerous tumour near his pituitary gland began affecting his eyesight. Doctors said the growth could eventually have caused him to lose his vision if left untreated.
The AI technology monitored a live video feed from inside Hibbert’s skull during the operation. It identified and highlighted the likely locations of critical blood vessels and nerves that were hidden from direct view, helping surgeons determine where it was safer to remove the tumour.
Hibbert had been fit and healthy until around 18 months ago, regularly taking a two-mile lunchtime walk while working as a customer services manager. During one of those walks, he suddenly collapsed and suffered a seizure, prompting doctors to carry out a brain scan.
The scan revealed an 11mm non-cancerous tumour growing on his pituitary gland, a small but crucial organ located at the base of the skull. The gland produces hormones involved in functions including growth, metabolism and the regulation of blood pressure and body temperature.
The tumour was positioned dangerously close to the carotid arteries, which supply blood to the brain, as well as the optic nerves responsible for vision. As the growth increased pressure on those nerves, Hibbert began losing part of his peripheral vision and later experienced dizziness, severe tiredness and difficulty avoiding objects while walking.
Despite the challenges, Hibbert said he was determined not to allow the condition to stop him from maintaining his normal family life. He said the emotional and psychological effects had been particularly difficult, including concerns about becoming a burden to those around him.
When doctors offered him the opportunity to become the first patient in the trial of their bespoke AI-assisted surgical system, Hibbert agreed without hesitation. He said medical advances depended on patients being willing to participate in research so doctors could continue learning.
The procedure involved passing a thin camera, known as an endoscope, through Hibbert’s nose and towards the base of his skull. Surgeons then had to reach the tumour, which was concealed behind layers of bone and tough membrane.
Finding the safest route was complicated because the precise arrangement of blood vessels and nerves differs from one patient to another. Surgeons also face significant risks during the procedure, including the possibility of leaving part of the tumour behind or damaging a major blood vessel.
The experimental AI system was displayed on a second screen during surgery. It tracked the surgical instruments and interpreted the live camera footage, marking areas where important vessels and nerves were most likely to be located and identifying regions considered safer for tumour removal.
Researchers trained the system using hundreds of recordings of pituitary tumour operations. Specialists manually outlined key blood vessels and nerves in the footage, allowing the AI to learn how to recognise anatomical structures that may not be directly visible during surgery.
Professor Hani Marcus, one of the neurosurgeons involved in the operation, said the technology had effectively been exposed to more surgical cases than most individual surgeons would encounter during their careers. He described it as an additional expert pair of eyes available to the operating team.
Unlike some experimental medical AI systems that analyse information before or after a procedure, the technology used in Hibbert’s operation works while surgery is taking place. However, the medical team stressed that the AI does not make decisions independently and that surgeons remain fully responsible for the procedure.
Researchers ultimately hope to develop an AI assistant resembling ChatGPT for surgeons, giving medical teams the option of seeking additional advice during complex procedures while retaining the ability to reject its recommendations.
The technology has been developed and tested in laboratories for several years with funding from the National Institute for Health and Care Research and Google. Following the successful first human operation, the researchers are now preparing for a larger clinical trial.
For Hibbert, the outcome has already been transformative. He said that when he woke up after surgery, his vision appeared dramatically clearer, describing the experience as though he had regained a full field of sight after struggling with impaired vision for about a year.
Eight weeks after the operation, he said his eyesight continued to improve, while his energy had returned and the symptoms that had troubled him before surgery had disappeared.
“It’s given me my life back,” Hibbert said.
Health Innovation Minister James Frith welcomed the procedure, describing Hibbert’s treatment as life-changing and highlighting the potential of government-supported research. He also stressed that the use of AI in healthcare must be accompanied by appropriate safeguards, while ensuring patients can benefit from advances in the technology.
