AI-PRED

  • Research type

    Research Study

  • Full title

    Investigating the use of AI-ECG in prediction of clinical outcomes in the Emergency Department

  • IRAS ID

    334866

  • Contact name

    Nicholas Peters

  • Contact email

    n.peters@imperial.ac.uk

  • Sponsor organisation

    Imperial College London

  • Duration of Study in the UK

    2 years, 0 months, 1 days

  • Research summary

    Heart diseases can lead to a condition where the heart struggles to pump blood as it should. This condition is known as Left Ventricular Systolic Dysfunction (LVSD) and can cause other serious health problems like heart failure, irregular heartbeats, and more. It's crucial to detect LVSD early on so that doctors can start treatment immediately.
    Currently, doctors often use a test called an echocardiogram to see how well the heart is pumping. But not every patient can get this test because it is expensive, and not all hospitals have the equipment or trained staff to do it, especially in busy places like the emergency department (ED).

    Recent developments in technology, specifically artificial intelligence (AI), have found a way to predict LVSD using a common heart test called an AI-ECG. Further investigating to see how reliable AI is in real-life emergency situations, where patients might have a mix of health issues.
    Using AI this way in the ED could:

    • Quickly identify heart patients and start treatment faster.
    • Make the ED process smoother, reducing long wait times.

    We are not entirely sure how this AI identify LVSD. Our study wants to see how well this AI approach works in the ED and what effects it has on patient care. Also, the AI-ECG can potentially identify patients who are at risk to develop heart disease. If it is successful, it could change the way we handle heart emergencies and lead to better, quicker treatments tailored to each patient's needs.

    The study will recruit patient attended at the ED. All patients will be consented and screened using a smart stethoscope called Eko DUO that has the ability to capture a single lead electrocardiogram powered by AI that could identify LVSD. Each participant will only be screened once without the need for any follow-up.

  • REC name

    London - Central Research Ethics Committee

  • REC reference

    24/PR/0228

  • Date of REC Opinion

    19 Apr 2024

  • REC opinion

    Further Information Favourable Opinion