A patient in the UK has become the first person in the world to undergo brain surgery using an artificial intelligence system that analysed the operation in real time, helping surgeons identify critical structures while removing a tumour.
Rhys Hibbert, a 48-year-old father of two from Bedfordshire, underwent the procedure at University College London Hospitals after doctors discovered a tumour growing on his pituitary gland. The non-cancerous growth had begun pressing against his optic nerves, affecting his peripheral vision and raising concerns that his sight could deteriorate further.
During the operation, surgeons remained fully in control, but an AI system provided an additional layer of information by analysing live video from the surgical camera. The technology identified where important blood vessels and nerves were likely to be located, including structures that were not immediately visible to the surgical team.
The aim was to help doctors remove as much of the tumour as safely possible while reducing the risk of damaging nearby structures.
Hibbert’s health problems began around 18 months before the surgery when he suddenly lost consciousness and suffered a seizure while taking his usual lunchtime walk.
A subsequent brain scan revealed an 11mm tumour on his pituitary gland, a small organ located at the base of the brain that produces hormones involved in functions including metabolism, growth, blood pressure and body temperature.
Its location makes surgery particularly delicate. The pituitary gland sits close to the carotid arteries, which supply blood to the brain, as well as the optic nerves responsible for vision.
Although Hibbert did not initially require an operation, his condition began affecting everyday life. He experienced severe tiredness and dizziness, started tripping over objects and noticed his peripheral vision becoming increasingly restricted.
When doctors offered him the opportunity to become the first patient to undergo the procedure with assistance from their experimental AI system, he agreed.
The operation was carried out using an endoscope, a thin instrument containing a camera, which was inserted through the nose to reach the base of the skull. Surgeons then had to navigate towards the tumour while avoiding major blood vessels and nerves.
Individual anatomy can vary considerably between patients, meaning scans taken before an operation do not always provide surgeons with a complete picture of what they will encounter during the procedure.
The AI system was designed to provide additional information at precisely this stage.
As the operation progressed, the technology analysed the live endoscopic video on a separate screen, followed the position of surgical instruments and highlighted areas where important anatomical structures were likely to be located.
Rather than replacing the surgeon’s judgement, the system effectively acted as an additional pair of eyes.
Researchers developed the technology by training and evaluating it using hundreds of videos from previous pituitary tumour operations. Important blood vessels and nerves were carefully labelled in the footage, allowing the system to learn patterns associated with structures surgeons need to identify and protect.
Professor Hani Marcus, one of the neurosurgeons involved in the operation, said the system had effectively been exposed to more procedures than many individual surgeons would see during their careers.
This could eventually become particularly useful in highly specialised procedures that surgeons may perform only a limited number of times each year.
The technology also points towards a potentially significant new direction for artificial intelligence in healthcare. Much of the discussion around medical AI has focused on analysing scans, assisting diagnoses and processing patient information. Using AI during an operation introduces the technology directly into one of medicine’s most complex environments.
The researchers behind the project ultimately envisage systems that could provide surgeons with expert assistance during procedures, offering information when required while leaving final decisions with the medical team.
That distinction will be important as AI becomes more deeply integrated into healthcare. A system capable of recognising anatomical structures can provide valuable information, but responsibility for interpreting that information and deciding how to proceed remains with trained medical professionals.
The technology was developed by researchers at University College London following years of laboratory testing, with funding from the National Institute for Health and Care Research and Google. Following the first procedure, researchers are now working towards a larger clinical trial to assess its effectiveness more widely.
For Hibbert, however, the impact has already been significant.
After waking from the operation, he noticed an immediate improvement in his vision. In the weeks that followed, he reported that his sight continued to improve while the fatigue and other symptoms he had experienced before surgery disappeared.
The successful procedure does not mean AI-assisted brain surgery is ready to become routine. Larger trials will be needed to establish how reliably the technology performs across different patients and surgical teams, while medical regulators will need to consider the safeguards required as AI becomes increasingly involved in clinical decisions.
But the operation provides an important glimpse of how artificial intelligence could develop inside hospitals.
Rather than a future in which machines replace surgeons, the more immediate possibility may be one in which doctors operate with increasingly sophisticated digital assistance, combining human experience and judgement with AI systems capable of analysing enormous amounts of medical information in real time.
For one patient in Britain, that future has already begun.
