Intelligent Energy-Aware Thermal Exchange Optimization with Deep Learning Model for IoT-Enabled Smart Healthcare

Area of Science:

• Computer Science
• Biomedical Engineering
• Artificial Intelligence

Background:

• The integration of Internet of Things (IoT) and advanced sensors is revolutionizing healthcare by connecting devices, patients, and professionals.
• Smart healthcare systems improve patient safety, accessibility, and operational efficiency while reducing costs.

Purpose of the Study:

• To develop an intelligent energy-aware thermal exchange optimization with deep learning (IEA-TEODL) model for IoT-enabled smart healthcare.
• To effectively cluster IoT devices and enhance disease diagnosis within the smart healthcare sector.

Main Methods:

• The proposed IEA-TEODL model utilizes an energy-aware chaotic thermal exchange optimization-based clustering (EACTEO-C) scheme for device grouping.
• Disease classification is performed using preprocessing, K-medoid clustering for outlier removal, Multihead Attention Bidirectional Long Short-Term Memory (MHA-BLSTM), and Weighted Salp Swarm Algorithm (WSSA) for hyperparameter tuning.

Main Results:

• The IEA-TEODL approach demonstrated enhanced classification outcomes through outlier removal and WSSA-based hyperparameter optimization.
• Extensive simulations on benchmark datasets confirmed the superior performance of the IEA-TEODL technique compared to existing methods across various evaluation metrics.

Conclusions:

• The developed IEA-TEODL model offers a significant advancement in IoT-enabled smart healthcare systems.
• The proposed technique effectively optimizes IoT device clustering and improves the accuracy of disease diagnosis, paving the way for more efficient and cost-effective healthcare solutions.