The Accelerated Capability Environment (ACE) and NHS AI Lab have collaborated to create an innovative AI tool designed to identify patients at risk of extended hospital stays, leading to improved patient outcomes and reduced healthcare costs.
This tool analyses hospital data using AI technology to predict which patients may require prolonged hospitalisation, allowing for early intervention and adjustments to treatment plans. Extended hospital stays can lead to higher mortality rates, readmission risks, and physical decline, especially in elderly patients.
Gloucestershire Hospitals NHS Foundation Trust, involved in the project, found that 4% of admissions resulted in stays of 21 days or longer, accounting for 34% of all bed stays.
The AI model, developed in partnership with Polygeist, was trained on 460,000 anonymised records and delivered a proof of concept within just 12 weeks. When tested at Gloucestershire Hospitals, the tool successfully detected 66% of patients in the highest-risk categories for extended stays.
Reducing the average hospital stay by even one day can result in significant cost savings, such as £1.7 million for Gloucestershire Hospitals. The tool was integrated with the trust's electronic health record system via application programming interfaces (APIs).
Similar AI initiatives are also being employed to identify high-risk patients on waiting lists for timely intervention.
Click here to read the original news story.