AI helps surgeons remove brain tumour in world-first operation

The world’s first patient to undergo brain surgery with live artificial intelligence assistance has successfully had his tumour removed, marking a major development in the use of AI in neurosurgery.

The world’s first patient to undergo brain surgery with live artificial intelligence assistance has successfully had his tumour removed, marking a major development in the use of AI in neurosurgery.

Rhys Hibbert, a 48-year-old father of two from Bedfordshire, underwent the pioneering procedure at University College London Hospital (UCLH). Surgeons said the tumour could have caused him to lose his sight if it continued to press on his optic nerves.

Hibbert’s tumour was an 11mm non-cancerous growth on the pituitary gland, a small organ located at the base of the skull. The gland produces important hormones involved in functions including growth, metabolism and the regulation of blood pressure and body temperature.

The tumour was located close to the carotid arteries, which supply blood to the brain, as well as the optic nerves responsible for vision. As the tumour grew, it began putting pressure on Hibbert’s optic nerves and affected his peripheral vision.

He had initially been fit and healthy but later experienced severe fatigue dizziness and difficulties walking. He eventually agreed to become the first patient to undergo the experimental operation using the bespoke AI system.

During the procedure, surgeons inserted a thin camera, known as an endoscope, through the nose to reach the base of the skull. The tumour was hidden behind layers of bone and a tough membrane making it difficult to identify the safest route for removal.

The AI system analysed a live video feed of the operation on a separate screen. It tracked surgical instruments and highlighted areas where critical blood vessels and nerves were likely to be located, helping surgeons identify safer areas from which to remove the tumour.

Researchers describe the technology as similar to facial-recognition systems used in smartphones, except that it is designed to recognise important anatomical structures that may not be immediately visible during surgery.

The AI was trained using hundreds of videos of pituitary tumour operations. Researchers manually identified blood vessels and nerves in the footage to help the system learn to recognise these structures.

Professor Hani Marcus, who helped perform the procedure, said the system had been trained on more operations than most surgeons would encounter during their careers and could provide an “expert second pair of eyes.”

However, the researchers stressed that the technology does not replace surgeons. Doctors remain fully responsible for decisions during the operation and can choose whether or not to follow the AI’s recommendations.

The team, which includes researchers at UCL and neurosurgeons from the National Hospital for Neurology and Neurosurgery, hopes to eventually develop an AI assistant similar to ChatGPT that surgeons can consult during complex procedures.

The technology was developed with support from the National Institute for Health and Care Research and Google after years of laboratory testing. Researchers are now preparing for a larger clinical trial.

For Hibbert, the results have been life-changing. He said his vision immediately felt dramatically clearer after waking from surgery and has continued to improve in the weeks since.

He has also regained his energy, while the symptoms he experienced before the operation have disappeared.

“It’s given me my life back,” Hibbert said.

The pioneering procedure could represent an important step toward using real-time AI assistance to improve safety and precision during complex brain surgeries.