Skip to the content

NHS Digital tests machine learning for hospitals’ Covid-19 response


Mark Say Managing Editor

Get UKAuthority News


Trials have begun of a system using machine learning to predict the approaching demand across England for intensive care beds and ventilators for patients with Covid-19.

NHS Digital said the Covid-19 Capacity Planning and Analysis System (CPAS) has been developed by its data scientists and researchers from the University of Cambridge, using data from Public Health England (PHE) and aimed at supporting hospitals in their planning.

It has been built on the Cambridge Adjutorium machine learning engine, developed by a Cambridge team led by Professor Mihaela van de Schaar and which has already been used to obtain insights on cardiovascular disease and cystic fibrosis. It is using data collected by PHE’s 19 Covid-19 Hospitalisation in England Surveillance System (CHESS).

Alpha stage trials have begun at four hospitals. They are aimed at ensuring that the ventilators, other equipment and drugs needed in intensive care units will be in place as they are needed.

It is also aimed at helping to predict the length of hospital stays and discharge planning.

Predicting demand

Professor Jonathan Benger, NHS Digital’s chief medical officer, said: “With the pressure being placed on intensive care by the current coronavirus pandemic it is essential to be able to predict demand for critical care beds, equipment and staff. CPAS allows individual hospitals to plan ahead, ensuring they can give the best care to every patient.”

CPAS provides hospitals with three types of information: statistics on patients being admitted with key data such as additional medical conditions; forecasts on the need for beds, ventilators and other resources; and a simulation environment to test alternative scenarios such as changes in the profile of patients admitted.

Dr Jem Rashbass, executive director for master registries and data at NHS Digital, said: “Two weeks ago, the team shared a method with the world that showed it was possible to do capacity planning for Covid-19 patients. 

“We recognised that there was an opportunity to industrialise the methods and deploy this as a service through the national infrastructure managed by NHSD and deliver a real data driven planning tool to hospitals.”

Professor van der Schaar emphasised that CPAS does not make treatment decisions about individual patients but aggregates the data for more accurate predictions about larger groups, at the level of a hospital, a trust, a region or nationally.

NHS Digital said that if the trials prove successful over the coming days the system will be rolled out across the NHS, and that several other countries have expressed an interest.

In the longer term it could be developed into a broader framework for managing hospital resources.

Image by Peter Stevens, CC BY 2.0 through flickr

Register For Alerts

Keep informed - Get the latest news about the use of technology, digital & data for the public good in your inbox from UKAuthority.