Exeter AI project secures £300k Government funding to detect hospital infections before symptoms appear

An artificial intelligence system developed in Exeter could soon help doctors detect life-threatening hospital infections up to a week before patients show any symptom, potentially transforming patient safety at the city’s main hospital and beyond.

An Exeter-led team has secured more than £300,000 in Government backing to advance the technology, with trials now under way at the Royal Devon University Healthcare NHS Foundation Trust.

The funding comes through an Innovate UK SMART grant, awarded to support high-potential, cutting-edge innovation. 

It will enable researchers and clinicians to expand and refine the AI platform, known as MEMORI, which analyses patient data in real time to predict the early risk of infection before visible signs appear.

The project is being delivered through a partnership between the Exeter-based NIHR HealthTech Research Centre in Sustainable Innovation and UK health technology firm Sanome, with support from the NIHR Exeter Biomedical Research Centre and the University of Exeter.

READ NEXT: Police issue public appeal for wanted man with links to Exeter and Exmouth

Hospital-acquired infections - including pneumonia, MRSA and C. difficile - remain one of the NHS’s most persistent and costly challenges. 

Nationally, they are linked to more than 7.1 million excess bed days every year and cost an estimated £2.7 billion. Research suggests that up to 55 per cent of such infections could be prevented if identified earlier.

For the Royal Devon, which treats around 180,000 inpatients and day cases annually and delivers more than 500,000 outpatient appointments each year, even a modest reduction in infection rates could have significant consequences. 

Fewer infections would not only improve patient outcomes but could also free up much-needed bed capacity in a system that regularly runs close to full occupancy, particularly during winter pressures.

Dr Nick Kennedy, Digital Innovation and AI Theme Lead at the NIHR HealthTech Research Centre in Sustainable Innovation and Consultant Gastroenterologist at the Royal Devon, said early detection is critical, particularly for vulnerable patients.

“Hospital-acquired infections remain one of the biggest threats to patient safety, particularly for vulnerable patients with complex conditions. That means early intervention is vital,” he said.

“By co-designing MEMORI with the support of Innovate UK, we are proud to be among the first to test such technology and show how AI can support clinicians, transform patient care and ultimately save lives.”

MEMORI analyses multiple streams of patient data - including observations, laboratory results and other clinical information - to identify patterns that may indicate infection risk. 

Early studies suggest it outperforms the NHS-standard National Early Warning Score (NEWS2) system in detecting deterioration, potentially giving clinicians a critical window of several days to intervene sooner.

The new funding will allow the Exeter team to expand the range of patient information the system can process, improving predictive accuracy. 

It will also support deeper integration with the Royal Devon’s electronic patient record systems, ensuring alerts can be delivered seamlessly into clinicians’ everyday workflows.

Benedikt von Thüngen, CEO and Founder of Sanome, said the project highlights how NHS data can be used responsibly to improve frontline care.

“Our mission is to prevent deterioration before it becomes life-threatening. MEMORI shows how real-world NHS data, when safely and securely unlocked, can be transformed into actionable bedside insights that change outcomes using the power of multimodal AI.”

Chris Sawyer, Innovation Lead for Digital Health at Innovate UK, described the partnership as a strong example of how clinical expertise and innovation can combine to tackle long-standing healthcare challenges.

The project further strengthens Exeter’s reputation as a centre for digital health research and innovation, with collaboration between NHS clinicians, university researchers and industry partners.

Early findings from the Royal Devon trial are expected later in 2026. If successful, the Exeter-developed system could be rolled out more widely across other NHS organisations.