A British man has become the first patient in the world to undergo brain surgery with live artificial intelligence assistance, with surgeons successfully removing his tumour.
Rhys Hibbert, a 48-year-old father of two from Bedfordshire, was at risk of losing his sight because the tumour was pressing against the nerves responsible for vision.
Doctors at University College London Hospital said the AI system provided real-time guidance during the operation, helping the surgical team identify important blood vessels and nerves that could not be seen directly.
Hibbert said the technology provided an additional layer of safety during a procedure where there are no clear visual markers to guide surgeons.
Tumour affected his vision
Hibbert had previously been fit and active, regularly walking about two miles during his lunch break. That changed when he suddenly lost consciousness and suffered a seizure while out walking.
A subsequent scan discovered an 11mm non-cancerous tumour on his pituitary gland, a small organ at the base of the brain that produces hormones involved in functions including growth, metabolism and blood pressure regulation.
The tumour was located close to the carotid arteries and optic nerves. As it grew, it began affecting Hibbert’s peripheral vision.
He also experienced severe tiredness, dizziness and problems with balance.
Despite the challenges, Hibbert said he was determined to maintain his independence and did not want his condition to prevent him from caring for his family.
When doctors offered him the opportunity to take part in the pioneering procedure, he agreed.
AI provides a second set of eyes
The operation was carried out through the nose using an endoscope, a thin camera that allows surgeons to reach the base of the skull without making a large opening in the head.
Removing the tumour can be difficult because critical blood vessels and nerves may be hidden behind bone and protective tissue. The anatomy also differs from one patient to another.
Researchers developed an AI system capable of analysing the operation’s video feed in real time. It tracks surgical instruments and identifies where important anatomical structures, including blood vessels and nerves, are likely to be located.
The technology was trained using hundreds of recordings from previous pituitary tumour operations. Researchers manually identified key structures in the footage, allowing the system to learn how to recognise them.
Professor Hani Marcus, one of the neurosurgeons involved, said the system had been trained using more operations than many surgeons would encounter during their entire careers.
The team stresses that the AI does not make decisions independently. Surgeons remain responsible for the operation and can accept or disregard the system’s recommendations.
Researchers hope the technology could eventually become a kind of “second expert” for surgeons, providing advice during complex procedures whenever required.
Patient reports major improvement
Hibbert says his recovery has been positive. After waking from the operation, he immediately noticed a significant improvement in his vision.
Eight weeks later, he said his eyesight continued to improve and that the severe fatigue and other symptoms he had experienced before the surgery had disappeared.
He described the operation as having given him his life back.
The project received funding from the National Institute for Health and Care Research and Google. Following the first successful use of the technology in a real operation, researchers are now preparing for a larger clinical trial.
Health innovation minister James Frith welcomed the development, while stressing that AI used in healthcare must be subject to strong safeguards.
The breakthrough could mark an important step in the use of real-time artificial intelligence during surgery, although further research will be needed before the technology can be used more widely.
Be the first to leave a comment