XBeats: A Real-Time Electrocardiogram Monitoring and Analysis System

This work presents XBeats, a novel platform for real-time electrocardiogram monitoring and analysis that uses edge computing machine learning early anomaly detection. The encompasses data acquisition ECG patch with 12 leads to collect heart signals, perform on-chip processing, transmit the healthcare providers in further analysis. provides dynamically configurable selection of active could be transmitted backend system. ranges from single lead complete 12-lead testing configuration. XBeats implements lightweight binary classifier detection reduce time action should abnormal conditions occur. initial phase is performed on (i.e., device paired patch) alerts can configured notify designated providers. Further deep full fidelity sent backend. A fully functional prototype has been implemented demonstrate feasibly usability proposed Performance evaluation shows achieve up 95.30% accuracy conditions, while maintaining high rate 441 samples per second. Moreover, analytical results energy consumption profile show 37 h continuous streaming.