This country launches an AI engine to predict the time of death of patients

The UK plans to launch an artificial intelligence (AI) tool to help doctors identify high-risk heart patients. This follows a study that found the AI ​​tool could accurately predict a person’s risk of death several years after a heart scan.

Referring from detikHealtha global research team led by Imperial College London has tested its AI model, AI-ECG risk estimation or AIRE against millions of electrocardiogram (ECG) results or devices for detecting heart attacks and other anomalies.

As a result, the model was able to predict a person’s potential death within ten years of the EKG and the results were 78 percent accurate. In addition, this device can also predict heart attacks, heart failure and heart rhythm problems.

Researchers say the system could be rolled out across the UK’s National Health Service (NHS) within the next five years. Trials with human patients are planned at several sites in London, and are expected to begin in mid-2025.

Researchers will also evaluate the benefits of the model using patients from outpatient clinics and hospital medical wards.

“We think this could have huge benefits for the NHS, and globally,” Dr Fu Siong Ng, a cardiac electrophysiology researcher at Imperial College London who worked on the project, said in a statement , reported by Euronews.

The Potential of AI to Improve Heart Health

As is known, AI-powered ECG has been used to diagnose heart disease. However, it is not yet part of standard medical care and has not been used to identify a specific patient’s risk level.

“This could take the use of ECGs beyond what was previously possible, by helping to assess the risk of future heart and health problems, as well as the risk of death,” said Bryan Williams, head scientific and medical affairs at the British Heart Foundation, which funded the research.

The researchers, who published their findings in the Lancet Digital Health journal, said the AI’s incorrect predictions could be caused by other unknown factors, such as whether the patient received treatment that more or did he die suddenly.

But they emphasized that the model can still capture small changes in the structure of the heart, which could be warning signs of disease or death but which doctors could miss.

“Cardiologists use our knowledge and common sense when looking at ECGs, sorting them into ‘normal’ and ‘abnormal’ patterns to help us find disease,” said Dr Arunashis Sau, an academic physician at Imperial College London who led the new research.

“However, AI models look for much more detailed details, so they can detect problems on EKGs that appear normal to us, and which may occur long before the disease is fully developed,” said Sau .

Sau said more research is needed in hospitals and other health care settings to determine what role the model will play in future diagnosis and treatment.

Ng said.

This article was published on detikHealth.

(juice)