As the NHS needs to effectively predict the amount of ICU beds, ventilators and other vital life-saving equipment needed during the Covid-19 outbreak, trials of a new machine learning system to do just that have begun.
This new system is called the COVID 19 Capacity Planning and Analysis system and was developed by data scientists from the NHS and a team of research from the University of Cambridge.
The system will be using data from Public Health England to determine which resources are need. This will help the NHS across the UK to be able to deliver better care during the pandemic.
What the first stage of trials entails
A first round of trials is essential to see how the system will actually fair in a hospital setting. CPAS will be trialled in four hospitals to determine the accuracy and to see where it needs to be fine-tuned.
According to NHS Digital Chief Medical Officer Professor Jonathan Benger, this system is crucial because of the current pressure created by the coronavirus pandemic.
He hopes that CPAS can also be used to predict the length of hospital stay for patients and predict wider intensive care demand after the pandemic.
How will CPAS work?
This system is built around a machine learning engine called Cambridge Adjutorium, a flexible system that was specifically created for medical researchers. It was designed by Professor Mihaela van der Schaar and her team at the University at Cambridge.
Her team worked with NHS Digital and used data collected by Public Health England’s COVID-10 Hospitalization in England Surveillance System (CHESS).
CPAS uses information based on three key elements:
– Statistics providing insight into the demographics of patients, as well as their medical conditions.
– Forecasts to help determine the need for beds and equipment.
– A so-called “simulation environment” which will enable planners to test alternative scenarios.
Professor van der Schaar stated in an NHS press release that the system doesn’t make decisions about treatment plans for individual patients, but that it would make accurate predictions based on the data – this would help hospitals manage their resources better, according to predictions stemming from accurate data.
What will happen next?
If the trial is successful, CPAS will be rolled out through the NHS on a wider base, as there will be a need in the future to manage resources better, even after the pandemic.
According to Professor van der Schaar, a project like this could create stronger digital infrastructure for healthcare systems, which could help save lives, during the pandemic and long after that.
Article source: https://digital.nhs.uk/news-and-events/news/trials-begin-of-machine-learning-system-to-help-hospitals-plan-and-manage-covid-19-treatment-resources-developed-by-nhs-digital-and-university-of-cambridge