A new artificial intelligence (AI) tool developed in the UK can quickly rule out heart attacks in people attending A&E, saving tens of thousands of unnecessary hospital stays every year, according to its creators. can help to avoid
Known as Rapid-RO, the AI tool successfully ruled out heart attacks in more than a third of patients during trials at four UK hospitals.
Professor James Lepper, Associate Medical Director of the British Heart Foundation (BHF), which funded the research, said: “This research shows that AI can play a vital role in guiding treatment decisions for heart patients. can
“By quickly identifying patients who are safe for discharge, this technology can help people avoid unnecessary hospital stays, freeing up valuable NHS time and resources. is where it can be of greatest benefit.”
Blood tests are usually used to confirm the diagnosis when a patient arrives at the hospital with a suspected heart attack.
These tests measure levels of a protein called troponin that increases when the heart muscle is damaged.
However, this increase may not be reliably seen until hours later so people are often hospitalized for further troponin testing and monitoring.
Some of these patients will eventually be discharged without the need for treatment after refusing to have a heart attack.
BHF-supported PhD student Dario Sescia at Imperial College London developed Rapid-RO to identify patients at very low risk of heart attack.
Rapid-RO was trained using data from over 60,000 patients across the UK and then tested on over 35,000 patients.
It works by combining data from an initial troponin blood test with other patient information collected during hospital admission, which is analyzed by an algorithm.
Patients are then identified as either being in the very low risk group for heart attack, or not.
Rapid-RO was able to successfully rule out a heart attack in 36% of patients, compared to 27% ruled out by troponin blood testing alone.
It was also found to be more accurate in identifying heart attack cases.
The troponin test missed four times more heart attacks (108 cases) than the AI tool (27 cases), the researchers said.
He added that it was effective regardless of ethnicity, gender and the patients' covid-19 disease.
Dr Amit Kora, Postgraduate Clinical Research Fellow in Cardiology at Imperial College London, said: “Current methods of ruling out a heart attack combine clinical diagnosis with blood tests measuring troponin, which is a marker of heart muscle damage. The one is a sign of blood.
“Many patients require multiple troponin tests to confirm they have not had a heart attack, resulting in longer hospital stays and increased costs.
“We developed an artificial intelligence-based model, using age, the first set of blood tests – including troponin – and other basic health information, to help doctors end heart attacks sooner than with current methods. , while maintaining high accuracy across different age groups and in patients with different health conditions.
“Our study shows how artificial intelligence can help doctors make more timely decisions about patient care, prevent unnecessary hospital stays, and maintain patient safety.”
As part of the next steps, the researchers hope to turn Rapid-RO into an app that doctors can use.
Professor Lepper said: “We look forward to further research to understand how Rapid-RO can be used in the future to speed up clinical decisions, improve patient treatment and care.”
The findings were presented at the British Cardiovascular Society conference in Manchester.