Superhuman AI-powered ECGs move a step closer to clinical use
AI detection of hidden heart signals in ECGs, developed with the support of our funding over many years, is a step closer to being rolled out in hospitals following the launch of a spinout company.
Research at Imperial College London’s National Heart and Lung Institute (NHLI) has produced a new generation of artificial intelligence models, capable of unlocking information hidden inside a routine electrocardiogram (ECG). The work, led by Professor Fu Siong Ng’s group, is now being taken forward into a spinout company called Cardiovolt.ai. The aim is for the AI technique to be deployed in hospitals and used to improve patient care.
‘Superhuman’ technology
An ECG, which takes 10 seconds to record, is one of the most common tests in medicine. It measures the electrical activity of the heart, is mainly used to check heart rate and rhythm, and has traditionally been used to diagnose only a handful of conditions including heart attacks and irregular heart rhythms (arrhythmias). But AI can pull far more information from an ECG, transforming it into a powerful diagnostic tool.
Dr Arunashis Sau, Chief Scientific Officer of Cardiovolt.ai, a lecturer at NHLI and cardiology registrar at Imperial College Healthcare NHS Trust, said: “We looked at the ECGs to see if we could do things that are superhuman. Not things that clinicians can already do, but things that no cardiologist, no matter how expert, can do.”
Digital signals
The AI models were developed using a data bank of more than 1.6 million ECGs from Brazil - each linked to a patient’s medical history - and a further several million ECGs from the United States. What the models are detecting, embedded within the electrical readings from the heart and invisible to the human eye, are signals that there are underlying disease processes underway.The team found that the models performed strongly in testing, reaching diagnostic accuracy of between 83 and 93 per cent for heart disease and 70 to 80 per cent for non-cardiovascular conditions such as diabetes and kidney disease.
The next steps, made possible by the spinout, are to bring the AI technology into widespread hospital use.
The Cardiovolt.ai team (L-R) Professor Fu Siong Ng (Chief Medical Officer), Dr Arunashis Sau (Chief Scientific Officer) Boroumand Zeidaabadi (Chief Executive Officer) and Dr Libor Pastika (Chief Technology Officer)
Professor Fu Siong Ng, Professor of Cardiology at Imperial College London and a consultant cardiologist at Chelsea Imperial College Healthcare NHS Trust, oversaw the research and serves as Cardiovolt.ai’s Chief Medical Officer. He said: “We have chosen this path because it is the one most likely to see the technology used in hospitals. We are the people who are happy to take the risk, and devote our time and energy to making that happen.”
The initial clinical focus of Cardiovolt.ai is on diagnosis - detecting conditions like heart failure or valve disease that may otherwise go unnoticed.
Professor Ng added “If someone comes in to have an ECG, we want to pick up underlying heart failure or valve disease that would never be picked up by a human doctor. If there is any suspicion, then we will do an echocardiogram to confirm it right away.”
Development has also been supported by Imperial’s enterprise ecosystem, including the AI SuperConnector accelerator and the 2025 AI & Robotics Track prize in its Venture Catalyst Challenge.
Exciting leap forward
Our director of research, Professor James Leiper, said: “This spinout is a prime example of how BHF-funded research can lay the foundations for technology which could make a real difference to patients’ lives.“ECGs have been used to assess the heart for more than a century, but artificial intelligence is a gamechanger which can take them far beyond what was previously possible. This remarkable ability to spot hidden signals can give people a read-out of their risk for future heart and health problems.
“As the spinout speeds this technology towards patient use, we look forward to seeing how it could accelerate decision-making for clinicians, so that patients can quickly receive the diagnoses and treatments that can help them to live healthier, longer lives.”