Trials have begun of a system that will use machine learning to help predict the upcoming demand for intensive care (ICU) beds and ventilators needed to treat patients with COVID-19 at individual hospitals and across regions in England.
The COVID 19 Capacity Planning and Analysis System (CPAS), developed by NHS Digital data scientists and a team of researchers from the University of Cambridge, and using data from Public Health England, will support hospitals to plan more accurately and help ensure that resources are deployed to best effect to support COVID-19 throughout the NHS. The first stage alpha trials began this week at four hospitals, aiming to demonstrate the relative accuracy of the system and fine tune it to best meet the needs of hospitals.
“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,“says NHS Digital Chief Medical Officer Professor Jonathan Benger. “CPAS allows individual hospitals to plan ahead, ensuring they can give the best care to every patient. At the same time, the wider NHS can ensure that the ventilators, other equipment and drugs that each intensive care unit will need are in place at exactly the time they are required. In the longer term, it is hoped that CPAS can be used to predict hospital length of hospital stay, discharge planning and wider intensive care demand in the time that will come after the pandemic.”
Based on data for COVID-19 patients admitted to hospitals, CPAS will support a hospital to predict how many patients may require admission to ICU, how many may require ventilators, and how long patients are likely to be in hospital or ICU. Taken together, these predictions will provide information to support the hospital to plan the resources they will need to look after those patients, make up shortfalls, or share excess capacity with other hospitals.
CPAS is built around a machine learning engine called Cambridge Adjutorium, developed by Professor Mihaela van der Schaar and her team at the University of Cambridge. A highly flexible machine learning system created for medical
researchers, Cambridge Adjutorium has already been used to develop insights into cardiovascular disease and cystic fibrosis.
“Professor van der Schaar’s lab at the University of Cambridge is recognised as the UK’s and possibly the world’s leading team creating new methods in machine learning applied to address real-world medical problems,” says Dr Jem Rashbass, Executive Director for Master Registries and Data at NHS Digital. “Professor Van Der Schaar is an engineer, and she leads a multidisciplinary team that develops, tests and delivers working new solutions to the really hard problems in medicine. 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.”
To apply the power of machine learning to the front line of the current pandemic, the Cambridge team worked with NHS Digital and first turned to depersonalised data collected by Public Health England’s COVID-19 Hospitalization in England Surveillance System (CHESS), which was used to train the Cambridge Adjutorium model, demonstrating that it could successfully predict hospital resource use accurately.
“Although the system uses data from individuals to build its models, the system does not make treatment decisions about individual patients. Rather, by aggregating that data we can make more accurate predictions about larger groups, at the level of a hospital, a trust, a region or nationally. So while we can say with a high level of confidence that 30 out of 40 ITU beds in a hospital will be occupied next week, we are not trying to predict which patients will be in them,” explains Professor van der Schaar.
NHS Digital is now working in partnership with the Cambridge team to turn that demonstration into a robust tool that can be used on a daily basis by hospitals across the country, and to improve the capability and accuracy of the system by integrating a wider range of data collected by NHS Digital alongside the CHESS data. The trial that launches today is the first step in that process.
“We’re thrilled to be able to work with NHS Digital and Public Health England on this unique and pioneering project,” says Prof van der Schaar, “which brings not only high-quality data and multiple integrated data sets, but access to the entire ecosystem and the ability to implement and distribute a solution to hospitals and trusts.”
The information and capabilities available to hospitals using the CPAS tool in the trial are made up of three key elements:
- statistics, which provide a demographic picture of the population of patients being admitted and key information such as additional medical conditions, and which allow comparisons between the specific local hospital data and the regional or national picture;
- forecasts, which will help predict the need for beds, ventilators or other resources over the coming days;
- a “simulation environment” which will allow planners to test the effect of alternative scenarios, such as increasing the number of available beds or changes in the profile of patients admitted.
If the trial deployments prove successful over the coming days, the CPAS system will be rolled out across the NHS, and several other countries have already expressed an interest. In the longer term, there is a potential opportunity to develop the system into a broader framework for helping hospitals manage their resources more generally.
“COVID-19 is a terrible disease that has torn families and friends apart, brought the UK to a virtual standstill, and upended the lives of countless healthcare workers. But if we can use projects like this to strengthen collaboration and lay the foundations for a stronger digital infrastructure in healthcare, we can emerge from this pandemic even stronger,” concludes Prof. van der Schaar.