IoT-Blockchain Integration for Secure Smart Healthcare: Neural-Enhanced Encryption and Hybrid Anomaly Detection

Abstract

Background: The application of the Internet of Things (IoT) in healthcare has allowed continuous patient surveillance, data acquisition, and analytics for better patient care. However, the amount of healthcare data generated by IoT devices has created several security and privacy concerns related to information’s confidentiality, integrity, and scalability in the health care system. Problem Statement: The current healthcare system lacks mechanisms to ensure data confidentiality, integrity, and effective monitoring of real-time patient statistics while providing efficient and effective healthcare services. In addition, the conventional security measures may not be sufficient to detect various forms of attacks in a real-time healthcare IoT environment. Purpose: This paper proposes an efficient and reliable healthcare framework that addresses several IoT-related security issues while ensuring confidentiality and integrity of health data and providing effective anomaly detection. Methods: The study presents a framework that uses a combination of IoT, blockchain, artificial intelligence, and neural-enhanced encryption to offer a more secure and reliable healthcare system. It utilizes neural networks to provide encryption and implement an anomaly detection model that uses a hybrid long short-term memory autoencoder and support vector machine detecting abnormal physiological observations in IoT-generated healthcare data. In addition, the framework uses a modified proof-of-authority consensus algorithm to ensure faster transaction validation and reliable processing in a permissioned blockchain network. The framework has been tested using simulated healthcare IoT networks with 150, 600, and 1200 nodes and a database with 1200 patient records. Results: The proposed framework achieved an anomaly detection accuracy of 98.99% with 0.89 precision, 0.93 recall, and 0.91 F1-Score. Besides, the framework detected 95% of the attacks and maintained optimal performance in the networks with varying node numbers. Conclusion: The study revealed that the combination of IoT, blockchain, intelligent anomaly detection, and neural-enhanced encryption could be an efficient and reliable solution to the current health care issues. The proposed framework offers several benefits, including improved confidentiality and integrity of health data, improved attack detection, and efficient and effective processing of transactions in a permissioned blockchain network.