Researchers at Imperial College London have developed an AI tool, Rapid-RO, that could decrease unnecessary hospital stays for A&E patients suspected of heart attacks. Rapid-RO analyses troponin blood test results and patient data to classify risk levels, allowing for quicker discharge of low-risk patients. Trials across four UK hospitals showed promising results, outperforming current testing methods. This innovative tool has the potential to optimise NHS resources and improve patient care.
A new artificial intelligence tool, developed by researchers at Imperial College London, could significantly reduce unnecessary hospital stays for patients in A&E with suspected heart attacks. The AI software, named Rapid-RO, was designed utilizing data from over 60,000 patients across the UK, incorporating age, blood test results, and other health information.
Rapid-RO functions by analyzing initial troponin blood test results alongside other collected data to classify patients as low or high risk for a heart attack, potentially allowing for quicker discharge of low-risk patients. In trials involving over 33,000 patients across four UK hospitals, the tool successfully ruled out heart attacks in 36% of cases, outperforming the 27% success rate of the current troponin testing method alone.
The development, led by Imperial College PhD student Dario Sesia and supported by the British Heart Foundation, indicates that the AI tool was effective across different ethnicities, sexes, and whether patients had Covid-19. Professor James Leiper from the British Heart Foundation highlighted the potential of this technology in optimizing NHS resources.
This new AI tool may offer a significant improvement in patient care and resource allocation within the NHS.
