London Surgeons Complete AI-Assisted Brain Tumour Removal

The AI system processed a live surgical video feed and color-coded structures including nerves and blood vessels near the tumour, according to The Guardian and The Independent. This provided visual guidance on anatomy that surgeons needed to avoid, while the surgical team remained in control of the operation, The Guardian reported.

The tumour was located on Hibbert's pituitary gland. Global News reported that the gland is surrounded by tightly packed blood vessels and vision-related nerves, leaving little margin for error. The Guardian, citing health officials, reported that an error of a millimetre can have serious consequences, including blindness, stroke, or death.

Professor Hani Marcus, a consultant neurosurgeon at the hospital and a University College London professor of neurosurgery, performed the operation, according to The Telegraph. The hospital had previously used the AI technology as a research tool, but this was its first use during surgery on a patient, The Guardian and The Independent reported.

This is a reported clinical milestone rather than evidence of comparative clinical performance. The available reporting describes one operation and does not provide accuracy metrics, false-positive rates, workflow latency, or outcomes compared with conventional image-guided surgery.

For clinical AI teams, real-time intraoperative systems introduce requirements beyond model development. Comparable systems generally require reliable video capture, low enough inference latency to fit surgical workflow, intelligible visualization, and validation across changing lighting, anatomy, instruments, and camera angles. They also require clear human oversight, particularly when an overlay can influence decisions in anatomy with limited tolerance for error.

The reported workflow used AI as decision support, not autonomous control. That distinction is material: the system highlighted anatomical structures, while surgeons made operative decisions. The reporting does not identify the model architecture, training dataset, regulatory status, or the broader study protocol.

Hibbert told The Guardian that he could see the room clearly when he awoke after surgery and was able to walk independently without glasses or sticks within a week. His experience is a positive individual outcome, but broader assessment of the technology will depend on results from additional patients and on transparent reporting of safety, usability, and clinical endpoints.