Neural Network Developed in Russia Can Detect Heart Arrhythmia in Real Time

A new Russian technology could give individuals the power to screen their heart health from a smartwatch, using AI to detect signs of arrhythmia before symptoms arise.

A team in St. Petersburg has developed a digital method to help detect heart arrhythmia early. The AI-driven system analyzes a person’s cardiac signals and compares them to a healthy heart rhythm. It provides a percentage score, where 100% indicates normal heart function.

The technology is lightweight enough to run on most smartwatches, removing the need for high-performance devices. But it's not just for personal use — doctors will be able to use the system in clinical settings, including intensive care units, where it can be integrated into Holter or ECG monitors to provide real-time cardiac analysis on a central display.

The tool was created by a student at the Saint Petersburg State University of Aerospace Instrumentation. The neural network has been trained to recognize a wide range of cardiac signals, including healthy rhythms, noisy patterns, exercise-induced signals, and those with arrhythmias. Its creators are now planning to partner with Russian medical companies to bring the technology into clinical practice nationwide.