Fog Computing-Enabled Adaptive Supervised Techniques for Real-Time ECG Data Management and Synchronization to the Cloud Based on IoT Devices

Electrocardiogram (ECG) data, characterized by its complexity and variability, presents significant challenges for real-time monitoring systems. The inherent processing in these systems can be exacerbated by the application of unsupervised learning techniques, which may struggle to interpret and process the nuanced patterns within ECG signals effectively. Our research identifies this processing issue and proposes a shift towards a supervised monitoring system, leveraging the expertise of clinical professionals. Moreover, we proposed a fog-based architecture for synchronizing real-time ECG data to the cloud. By incorporating supervised learning models and fog-based architecture, we aim to enhance the accuracy, responsiveness, and synchronization of ECG data interpretation, thereby reducing processing time and response time especially in case of critical condition of the patient during real-time monitoring. This article explores the comparative performance of supervised versus unsupervised learning techniques in ECG real-time monitoring, highlighting the benefits of clinical expert-driven supervision in achieving more timely and reliable results during real-time monitoring. Moreover, results showed that our fog-based synchronization approach performed better in terms of synchronization and processing time, and utilized bandwidth for data transmission.