London: A 48-year-old father-of-two has become the first patient to undergo brain surgery with live artificial intelligence assistance, with surgeons successfully removing his tumour while using the technology to identify critical blood vessels and nerves.
Rhys Hibbert, from Bedfordshire, underwent the pioneering procedure at University College London Hospital (UCLH) after a non-cancerous tumour began affecting his vision.
AI acts as a second pair of eyes
The bespoke AI system analysed a live video feed from the operation and tracked surgical instruments while highlighting areas where important blood vessels and nerves were likely to be located.
The technology was designed to help surgeons identify safer areas from which to remove the tumour, which was positioned close to the carotid arteries and optic nerves.
Prof Hani Marcus, one of the surgeons involved, said the AI had been trained using hundreds of videos of pituitary tumour operations and could function as an expert additional pair of eyes during surgery.
However, the medical team stressed that the AI did not make decisions independently. Surgeons remained in complete control and could choose whether to follow its guidance.
Tumour threatened patient’s eyesight
Hibbert was diagnosed after collapsing and suffering a seizure during a walk. Scans revealed an 11mm non-cancerous tumour on his pituitary gland at the base of the skull.
As the tumour pressed against his optic nerves, his peripheral vision began to narrow. He also experienced severe fatigue, dizziness and problems with balance.
The operation involved inserting an endoscope through his nose to reach the base of the skull. The tumour was located behind layers of bone and membrane, making it difficult to determine the safest route for removal.
Patient says surgery gave him his life back
Hibbert said his vision appeared dramatically clearer when he woke after the operation.
In the eight weeks following surgery, he reported continued improvement in his eyesight, along with the return of his energy and disappearance of the symptoms he had experienced before the procedure.
The research team, developed by UCL researchers and supported by the National Institute for Health and Care Research and Google, now plans to conduct a larger trial.
Researchers hope the technology could eventually develop into a form of “ChatGPT for surgeons”, providing doctors with an additional source of real-time guidance during complex procedures.