Reconceiving the Edge Intelligence Based IoT Devices for an effective Classification of ECG Systems

The rise of Internet of Things (IoT)-enabled healthcare systems has led to the development of real-time, intelligent solutions for medical data processing, especially in electrocardiogram (ECG) signal classification. As wearable IoT devices for health monitoring become more widespread, there is increasing demand for an effective method that can run on edge devices, ensuring low-latency processing and real-time decision-making. This research introduces a novel method that integrates Convolutional Neural Networks (CNN) with Long Short-Term Memory (LSTM) networks for precise ECG classification. The model utilizes CNNs for spatial feature extraction from raw ECG signals, identifying key patterns like P-waves, QRS complexes, and T-waves, which are essential for accurate classification but challenging to extract manually. To capture the sequential nature of ECG signals, the model incorporates LSTM layers, which are effective at retaining long-range dependencies and recognizing patterns indicative of cardiovascular conditions. The system is trained and validated using ECG data collected from IoT-enabled wearable sensors, ensuring real-world applicability in edge computing environments. The model is designed to handle the constraints of edge devices, such as limited computational power, while maintaining high classification accuracy. The hybrid CNN-LSTM model achieves a 99% accuracy rate, surpassing existing Machine Learning (ML) models in terms of sensitivity, specificity, and overall performance. This approach offers a promising direction for integrating AI-based analytics into IoT-driven healthcare systems, enabling real-time, accurate decision-making for early diagnosis and intervention. It enhances IoT healthcare systems' scalability and practicality, improving patient monitoring and cardiovascular health outcomes.