A Survey of Methods and Architectures for Smart Healthcare Patient Monitoring System for IoT-Based Healthcare System Using Enhanced Residual Multi-Scale Diverged Self-Attention Network

Abstract

Smart healthcare systems have transformed modern medical practices through the integration of Internet of Things (IoT) and Artificial Intelligence (AI) technologies. IoT-based healthcare monitoring systems enable continuous collection of physiological data such as heart rate, ECG, oxygen saturation, and body temperature using wearable devices and sensors. These systems generate large volumes of heterogeneous and high-dimensional data, requiring advanced analytical models for accurate prediction and diagnosis. Deep learning models such as Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and attention-based architectures have significantly improved the performance of patient monitoring systems. Hybrid models like CNN-LSTM effectively capture both spatial and temporal dependencies in biomedical signals, enhancing prediction accuracy and anomaly detection.  Recent advancements in self-attention mechanisms and Transformer architectures enable models to focus on relevant features and capture long-range dependencies in time-series healthcare data. Additionally, multi-scale and residual learning techniques improve feature extraction and model efficiency. The proposed Enhanced Residual Multi-Scale Diverged Self-Attention Network integrates these concepts to achieve superior performance in smart healthcare monitoring systems. Despite these advancements, challenges such as data heterogeneity, privacy concerns, computational complexity, and energy constraints remain. Emerging solutions such as edge computing and federated learning are being explored to address these issues. This survey reviews recent developments (2020–2023), identifies trends, and highlights future research directions in IoT-enabled smart healthcare monitoring systems.