In use from mid-2025 AI model to be able to predict time of death

Scientists at Imperial College London have developed an AI-based model that can predict health risks and potential lifespan. But the approach is not without controversy.

For years, scientists around the world have been researching methods to predict life expectancy. This is being done with the help of artificial intelligence.

Researchers at Imperial College London have developed a new AI-based model called AIRE (AI-ECG risk estimator), which uses electrocardiograms (ECGs) to predict health risks and the potential lifespan of patients.

An accuracy of 78 percent

The model was fed with over 1.1 million ECGs from almost 190,000 patients. According to the researchers, it can predict various health risks, heart disease and even if someone would die from non-heart-related causes with an accuracy of 78 percent.

"ECGs capture a lot of information from the whole body, because diseases such as diabetes, which affect organs such as the kidneys or liver, also affect the heart in some way," said the study authors.

The analysis therefore shows that AI can not only tell us a lot about the heart, but also about what is going on elsewhere in the body, and that it could be able to detect accelerated ageing.

UK health authority plans deployment from mid-2025

The NHS is planning to test the technology in selected clinics from mid-2025. The aim is to identify high-risk patients at an early stage and initiate appropriate measures.

The use of such AI models to predict health risks and life expectancy is the subject of lively debate in medicine. The researchers concede that the handling of such sensitive data raises ethical questions and that appropriate regulations are required.