Brain surgery is a risky business. Straying a millimetre in the wrong direction can be fatal to a patient. When removing a tumour, for example, if you go too small you risk not removing all of the tumour, whereas if you go too large you risk damaging critical structures in the brain. In the brain every tiny blood vessel is important because it may be supplying something vital – eyesight, body movement, bowel function, in fact virtually every system in the body.
AI systems can analyse hundreds of videos of brain surgery, and help surgeons to decide where the boundary lies between going too small and going too large. In this way AI systems can develop a level of experience it may take a surgeon many years to obtain.
It is in the light of such facts that the UK's Science Minister recently pledged £13m for AI research in healthcare during a visit to University College London (UCL). UCL is a world leader in AI and healthcare, and has been awarded more than £2.2m for four AI driven research projects relating to current health challenges.
During the visit, in a mock operating room, Dr Sophia Bano, Assistant Professor in Robotics and Artificial Intelligence at UCL, explained how her research is using AI to help advance endoscopy technology in pituitary surgery, assisting surgeons as they perform delicate keyhole surgery to remove small tumours from inside the skull.
In fact, UCL already has a start-up called Oxford Heartbeat, which has produced an AI software tool currently being used by surgeons in hospitals in the UK, Germany and Finland. It allows them to simulate and rehearse the placement of a stent inside a patient’s blood vessel, which is a highly intricate procedure, calling for great accuracy. Indeed, around 20% of brain stent surgeries are unsuccessful and require further surgical intervention.
The value of IP and start-ups to UCL can be seen by visiting the Higher Education Statistics Agency. According to this website, for the academic year 2021/22 the income generated from IP was £5.253m, and the income generated from the sale of shares in spin-offs was £6.845m, creating a total income in excess of £12m.