Enhancing healthcare classification with hybrid multimedia data processing and deep learning TNBO FCNN approach in...

Area of Science:

• Healthcare Technology
• Artificial Intelligence
• Data Security

Background:

• The proliferation of Internet of Things (IoT) in healthcare generates vast multimodal patient data, posing challenges for accurate classification, real-time processing, and data security.
• Existing machine learning models struggle with feature extraction, computational demands, and privacy in decentralized, resource-constrained healthcare environments.
• Lack of robust routing and data integrity mechanisms impedes the deployment of current solutions in real-world healthcare applications.

Purpose of the Study:

• To develop a hybrid deep learning framework for efficient and secure classification of multimodal healthcare data from IoT devices.
• To enhance data routing, hyper-parameter tuning, and privacy protection in decentralized healthcare settings.
• To optimize the framework for energy efficiency, scalability, and real-time monitoring in telemedicine and smart healthcare.

Main Methods:

• Integration of Tunable Nonlinear Bayesian Optimisation (TNBO) for efficient routing and hyper-parameter optimization.
• Utilization of a Fully Connected Neural Network (FCNN) for robust classification of multimodal patient data.
• Embedding blockchain technology to ensure data security, transparency, and immutability in decentralized IoT environments.

Main Results:

• The proposed TNBO + FCNN model achieved high performance metrics: 0.924 accuracy, 0.921 sensitivity, and 0.926 specificity.
• Outperformed existing methods with an F1-score of 0.915 and an AUC-ROC of 0.950.
• Demonstrated superior energy efficiency, scalability, and suitability for real-time healthcare monitoring.

Conclusions:

• The hybrid deep learning framework effectively addresses challenges in IoT-based healthcare data classification, security, and real-time processing.
• The integration of TNBO and blockchain technology offers a robust solution for secure and efficient multimodal data analysis in smart healthcare.
• The model's validated efficacy positions it as a promising advancement for telemedicine and real-time patient monitoring systems.