An artificial intelligence model can predict a person's risk of having a future stroke using only a 10-second electrocardiogram, or ECG, according to Labmate-Online.
An ECG is a quick, painless test that records the heart's electrical activity, usually through small sensors placed on the skin. It is already one of the most common tools in medicine, used routinely in clinics and hospitals. What's new here is the idea that an AI system can read patterns in that brief recording and translate them into a forecast of stroke risk — something a clinician glancing at the same trace would not necessarily see.
Labmate-Online reports that the model works from a single, standard 10-second test, which is notable because it suggests the approach could fit into existing workflows rather than requiring new or expensive equipment. The appeal of such a tool is speed and simplicity: if a routine test already being performed can also surface a warning sign, that information comes at little added cost or burden to the patient.
The available source does not provide details on how the model was trained, how accurate it is, the size of any study behind it, or whether it has been approved for clinical use. Those are important caveats, and readers should treat the finding as an early signal rather than a tool available at the doctor's office today.
Why it matters: stroke is a leading cause of death and long-term disability, and much of its damage is preventable when risk is caught early — so a way to flag danger from a test millions of people already take could help doctors intervene before a stroke ever happens.