AI ECG tool flags heart disease in under two seconds
Doctors have developed an AI tool that can detect signs of heart disease from a standard ECG in under two seconds, potentially speeding up diagnosis for millions of patients. Unlike a conventional ECG, which records the heart's electrical activity but cannot itself detect heart disease, the new tool extracts information beyond what a human can see to flag heart failure and heart valve disease, prompting patients to be fast-tracked for a confirmatory echocardiogram rather than facing lengthy waits.
The technology, trained on millions of ECGs, was presented at the European Society of Cardiology congress in Munich. In a US trial of 67,000 patients, funded by the British Heart Foundation, it correctly identified up to 81% of people with heart failure and up to 90% of those with heart valve disease. Researchers, including Prof Fu Siong Ng of Imperial College London, said the tool could also be used to opportunistically screen ECGs taken for other reasons, and future plans include developing handheld AI-led ECG readers for clinicians.
- New AI tool detects heart disease signs from ECGs in under two seconds
- Trial of 67,000 patients found up to 90% accuracy for valve disease
- Aim is to fast-track high-risk patients for scans and treatment
New here? Start with this
An electrocardiogram, or ECG, records the electrical signals that control the heartbeat. It is commonly used to check heart rhythm, but diagnosing problems such as heart failure or faulty heart valves often requires an echocardiogram, which uses ultrasound to show the heart’s structure and how it works.
Heart failure means the heart cannot pump blood around the body as well as it should. Heart valve disease affects the valves that direct blood through the heart; both conditions can be difficult to spot early, and an echocardiogram may not be immediately available.
Researchers are exploring whether patterns in ECG recordings can help identify people who need further checks. The approach could matter because ECGs are already taken for many reasons, though an AI flag would point to the need for a confirmatory test rather than provide a diagnosis on its own.
Both sides, in good faith
The strongest fair case each way — we don't pick a winner.
The case for
The tool could help clinicians identify heart failure and valve disease sooner, using ECGs patients already receive and directing those at higher risk to confirmatory scans. Faster triage may reduce delays and make better use of limited specialist capacity, while the reported trial results offer a promising basis for further evaluation.
The case against
An algorithm that misses some cases or flags people who do not have disease could create false reassurance, anxiety and extra demand for confirmatory scans. Before wider use, clinicians and health services may reasonably want evidence that it performs reliably across different patient groups and settings, and that the benefits outweigh the costs and risks of relying on AI screening.
AI Art Culture Health Medicine Technology
Read the full article at the source →
Originally published by The Guardian as “‘Superhuman’ AI tool spots heart disease in less than 2 seconds”.