Survey of Blockchain-Based Healthcare Data Security Using Quaternion and Encoder-Elliptic Curve Deep Neural Networks

Abstract

The rapid digital transformation of healthcare has generated vast volumes of sensitive patient information, including electronic health records, medical images, genomic data, physiological signals, and pharmaceutical records, while exposing critical security challenges beyond the capabilities of traditional cryptographic and access control mechanisms. This survey comprehensively reviews advanced healthcare security architectures integrating Quaternion-Based Evolutionary Gravitational Neocognitron Neural Networks (QEGNNN), Encoder-Elliptic Curve Deep Neural Networks (E-ECDNN), and blockchain technology into a unified security framework. QEGNNN exploits quaternion algebra and evolutionary optimization to improve feature representation while reducing computational complexity for multimodal medical data, including MRI, ECG, histopathology, and ophthalmic images. E-ECDNN embeds elliptic curve cryptography within deep encoder networks, providing lightweight yet robust security for resource-constrained healthcare devices such as wearable sensors and implantable monitors. Blockchain technology ensures decentralized, immutable, and transparent data management while supporting compliance with healthcare regulations including HIPAA, GDPR, and the European Health Data Space. Additionally, the Evolutionary Gravitational Search Algorithm jointly optimizes neural network parameters, cryptographic configurations, and blockchain consensus mechanisms. Based on twenty-five recent studies covering electronic health records, medical imaging, genomics, and federated learning, this survey highlights the effectiveness of integrated QEGNNN-E-ECDNN-blockchain frameworks and discusses future directions involving post-quantum cryptography, zero-knowledge proofs, and quaternion homomorphic encryption.