Google Deepmind to build AI assisted apps for Imperial College Healthcare
January 5, 2017
Google artificial intelligence subsidiary Deepmind has agreed a five year partnership with central London health care provider Imperial College Healthcare NHS Trust. Deepmind will work with Imperial College Healthcare to develop mobile clinical apps supported by machine learning and deep neural nets.
Imperial College Healthcare, centred at St Mary's Hospital Paddington and covering the Westminster area of central London, becomes the second NHS partnership for clinical apps for Deepmind. In November, Deepmind announced a similar agreement with the Royal Free, a health care provider covering the Hampstead and Highgate areas of London.
Over the last two years, Imperial College Healthcare has moved from paper to electronic patient records. Imperial College believes that mobile technology is the natural next stage of this work.
By giving clinicians access to health care apps that link to electronic patient records, doctors and nurses at Imperial College Healthcare should be able to access information on the move, react faster in response to changing patient needs, and provide better care.
Deepmind will be working with Imperial College Healthcare to deploy its clinical app, Streams. This supports clinicians in caring for patients at risk of deterioration, particularly with conditions where early intervention can make all the difference. The technology will notify nurses and doctors immediately when test results show a patient is at risk of becoming seriously ill. It will also enable clinicians at the Trust to securely assign and communicate about clinical tasks, and give them the information they need to make diagnoses and decisions.
In addition to St Mary's Hospital Paddington, Imperial College Healthcare also runs a number of other hospitals across west central London, including the country's tertiary specialist centre for cancer, the Hammersmith Hospital, Queen Charlotte's Hospital, and the Charing Cross Hospital. Specialists in the trust move between all these hospitals, and others within the Imperial College Healthcare group. A specialist can often be off-site when a patient's condition requires special and immediate attention.
The task management features in Streams are underpinned by a programme of widely published research and early product development carried out at Imperial College London University and Imperial College Healthcare, as part of an app called Hark co-founded by Lord Ara Darzi.
Using simulated data, Hark was found to improve the quality of transfer of information between staff and was rated by users as more effective and efficient, and less distracting than pagers. Hark became part of DeepMind in early 2016.
Deepmind will be implementing open and interoperable standards via what’s known as a FHIR API. This should allow the Trust to easily, securely and consistently integrate other apps that could improve care, whether developed by third parties or innovators within Imperial College Healthcare.
“Apps have changed the way we live our lives, from banking to shopping, and they are clearly part of the future healthcare landscape," said Sanjay Gautama, Caldicott Guardian and Chief Clinical Information Officer at Imperial College Healthcare NHS Trust. "They bring immense opportunities for faster and more efficient care, by making access to vital information quicker and easier for clinicians. But for apps to be useful and safe they cannot operate in isolation – they need to be securely linked to the core electronic patient record system".
“By working with DeepMind we are embracing the opportunities that technology brings to improve patient care, using their expertise to help us deploy a system that allows us to maximise future innovations in mobile technology for healthcare for the benefit of our patients.”
Professor Ara Darzi, Director of the Institute of Global Health Innovation at Imperial College London, and co-inventor of the Hark app, said "Through the NIHR-funded Imperial Patient Safety and Translational Research Centre (PSTRC), a leading team of patient safety researchers have demonstrated how a failure to provide clinicians with the right information at the right time leads to safety breakdowns and poorer outcomes for patients. Working with DeepMind we have been able to translate cutting edge research insights from our Hark research programme into the feature set of the Streams platform. We will be measuring what impact Streams has on patient safety and clinical outcomes over the next couple of years.”