A Wearable Sensor Node For Detecting Atrial Fibrillation Using Real-Time Digital Signal Processing

Atrial fibrillation (AF) is the most common chronic arrhythmia affecting nearly nine million people in the European Union only. Ischemic stroke is the most feared condition caused by AF with a 4.8 times higher risk of occurrence among AF patients compared to rest of population. The asymptomatic nature of AF causes many people to be left untreated. Previous studies have shown that detecting AF using mechanocardiography, e.g. recording mechanical movement caused by cardiac activity with motion sensors, is possible. We propose a custom Bluetooth enabled wearable device for measuring mechanocardiographic signals. On-board real-time signal processing is applied in order to determine whether the patient is suffering from AF. Pre-clinical measurement results suggest that the algorithm is able to distinguish between AF and normal sinus rhythm.