Cloud-based ECG monitoring using event-driven ECG acquisition and machine learning techniques.

Area of Science:

• Biomedical Engineering
• Cardiology
• Signal Processing

Background:

• Chronic heart disorders pose a significant global health challenge.
• Electrocardiogram (ECG) signals are crucial for diagnosing cardiac conditions.
• Current methods for ECG analysis can be computationally intensive and require substantial bandwidth.

Purpose of the Study:

• To develop an intelligent event-driven system for real-time ECG signal acquisition, compression, processing, and transmission.
• To enable efficient and accurate detection of chronic heart disorders.
• To provide a computationally efficient solution for automated cardiac arrhythmia diagnosis supporting mobile health monitoring.

Main Methods:

• Implementation of an intelligent event-driven ECG signal acquisition system.
• Real-time data compression and signal processing techniques.
• Post-processing including denoising, feature extraction, and classification algorithms.
• Evaluation of system performance using classification accuracy, F-measure, AUC, and Kappa statistics.

Main Results:

• Achieved an overall 2.6 times compression and bandwidth utilization gain compared to classical methods.
• Demonstrated a significant reduction in complexity and execution time for denoising, feature extraction, and classification.
• Attained a best classification accuracy of 94.07% for cardiac arrhythmia detection.

Conclusions:

• The designed event-driven solution offers a computationally efficient approach for automatic cardiac arrhythmia diagnosis.
• The system provides high-precision decision support for cloud-based mobile health monitoring.
• The event-driven nature significantly enhances compression and processing efficiency for ECG signals.