Skip to content
left end
left end
right end

SMARTT Critical Care Pathways

Safe, Machine Assisted, Real Time Transfer. An artificial intelligence based decision support tool to enable safer and more timely critical care transfer 

Chief Investigator



Funding Stream

Grant Ref


Dr Chris Bourdeaux

University Hospitals Bristol and Weston NHS Foundation Trust 

01/01/2021 - 31/12/2023 (36 months)

NIHR Invention for Innovation (i4i) Artificial Intelligence Award




The NHS is under strain and more people than ever need intensive care. This research will help us make better use of our precious critical care beds. We will develop and test an artificial intelligence tool that will help us work out who is well enough to leave intensive care. Our tool will free-up bed spaces faster and ensure we keep patients in intensive care until they are well enough to leave. This will improve the safety and health of patients and help hospitals run more efficiently. Intensive care doctors make hundreds of difficult decisions every day. Deciding if someone can leave intensive care is just one of these critical decisions. Unfortunately, doctors can t always make this decision right away, and they often get it wrong due to all the other demands on their time and brain power. This means around a quarter of patients stay in intensive care for too long or not long enough. Monitors and blood tests provide thousands of pieces of information about each patient. Over several years we have formed a team of doctors and scientists to develop artificial intelligence technology that takes this information and works out when patients are ready to be discharged. We have shown that this method performs better than humans can. In this project we plan to improve this artificial intelligence solution and test it in a real intensive care unit. We will create a tool for hospital staff that displays a constantly updated and reliable list of patients who are fit to be discharged. Artificial intelligence is very new in hospitals and we need to understand how patients feel about doctors and nurses using it to look after them. Patients may worry about computers having their health information and making decisions about their care. To ensure that our work is people-focused, we will set up a panel of patients to oversee the project and provide feedback at all stages. Some doctors and nurses may also be worried about a change to their practice. We will work with them to make sure the technology is well-suited to their working environments. This technology will improve the use of intensive care beds, help NHS staff in their high-pressured work and save lives. Starting with two local NHS trusts, our ultimate goal is to get this tool to doctors and nurses in intensive care wards across the country.