How King’s College London is using AI to treat cancer

Vaughn Highfield October 11, 2018

The NHS faces a near-insurmountable problem. The number of patients coming to use its services is increasing, while the number of qualified workers needed to meet that demand is dropping. Throw in some budget cuts and a government that seems to want to squeeze the service dry, the NHS finds itself in a rather sticky situation – even if it’s supposed to receive a cash injection for digital transformation.

How King’s College London is using AI to treat cancer

For King’s College London, the solution to this problem is turning to artificial intelligence. This doesn’t mean you’ll suddenly find yourself being treated by robots wielding scalpels or operating scanners. Instead, AI will be used to bolster an existing workforce of trained medical professionals.

Working in tandem with Nvidia and it’s Project Clara healthcare platform for autonomous devices, King’s College is looking to revolutionise the radiotherapy industry and treatment process.

As it stands today, radiologists have almost no time at all to work with their patients. According to Kimberly Powell, VP of Healthcare at Nvidia, radiologists used to have 20 seconds per image scan, per patient to help analyse a potential problem. Now they have less than five. That 15-second discrepancy could be more than enough for a radiologist to miss something potentially vital to a patient’s condition.

By using AI to interpret the data taken from medical devices used during treatment and diagnosis, radiologists can suddenly spend longer looking at that data. They can delve deeper and the AI can assist by highlighting areas it believes are troublesome or abnormal. Now radiologists can know that, even though they may not have been able to spend 20 seconds on an image, the information they’ve gleaned from it is far more advanced and helpful that they don’t need to spend 20 seconds staring at a screen to understand what’s going on.

“This is a huge opportunity to transform patient outcomes by applying the extraordinary capabilities of AI to ultimately make diagnoses earlier and more accurately than in the past,” said Professor Sebastien Ourselin, head of the School of Biomedical Engineering and Imaging Sciences at KCL. “This partnership will combine our expertise in medical imaging and health records with NVIDIA’s technology to improve patient care across the U.K.”

As part of the project, both Nvidia and King’s hope to crack the problem of federated learning by tapping into KCL’s clinical network. The hope is that, in time, the teams can work together to utilise anonymised patient data around the world to create an ever-evolving database for the AI to tap into, so it can become even better at detecting cancers earlier than existing methods.

Nvidia will co-locate its engineers with KCL researchers along with clinicians from major London hospitals. The programme will work with King’s College Hospital, Guy’s and St Thomas’, and South London and Maudsley and aims to plug into existing devices and systems to help for an easier transition to utilising patient data to improve cancer treatments across the UK and around the world.