AI model estimates heart function non-invasively
2026-08-09
Heart function, not hospital hardware, takes center stage in a new experiment in cardiology. An artificial intelligence model developed under the leadership of Ateneo de Manila University estimates how well the heart pumps blood using only non-invasive physiological measurements, according to a report by Danika Geronimo of the Ateneo Research Communications Department.
This project suggests that precise insight into left ventricular performance does not always require echocardiography or other complex imaging modalities, as long as algorithms can learn the subtle signatures hidden in routine signals such as blood pressure and heart rate variability. Researchers trained the system on paired datasets that linked these surface readings with established measures of cardiac output and ejection fraction, allowing the model to infer myocardial contractility from data that never pierce the skin.
The promise here is pragmatic rather than flashy. By relying on inexpensive sensors and standard monitoring equipment, the approach could expand access to functional cardiac assessment in community clinics and resource-limited hospitals that lack advanced cardiology infrastructure. It also hints at continuous monitoring scenarios, where changes in stroke volume or diastolic function could be flagged early through passive data collection, long before a patient is referred for high-end imaging.
Loading...
