Blockchain Enabled Failure Prediction And Severity Classification Framework For Secured Electronic Health Record

Abstract

Electronic Health Monitoring Systems (EHMS) play a vital role in modern healthcare newlineby enabling continuous patient monitoring through wearable and ambient sensors. newlineThese systems help in early detection of health deterioration, remote supervision, and newlinetimely clinical interventions. However, their increasing dependence on digital newlineinfrastructure exposes them to major cybersecurity risks such as ransomware attacks, newlineunauthorized access, and unexpected system failures. Such incidents can compromise newlinepatient safety, delay treatment, and erode trust in healthcare technology. Traditional newlinesecurity and intrusion detection systems, often rule-based or classical machine newlinelearning (ML)-driven, struggle to predict system failures, identify rare attacks, or newlineprovide transparent decision-making required for real-time hospital operations. newlineTo address these challenges, this research proposes two algorithm-driven newlineframeworks. The first framework integrates Multi-Layer Lagrange Maxout newlinePerceptron (ML-LMP) with Ethereum-Hyperledger blockchain to enable secure and newlineefficient Electronic Health Record (EHR) transmission. The framework combines newlineElliptic CryptGenRandom Curve Cryptography (E-CGR-CC) encryption, Roundedsquare Root Log Hash Algorithm-256 (RRLHA-256) for hashing, and Generalized-Pi newlineFuzzy Proof of Stack Consensus Rule (GPF-PoSCR) consensus-based smart contracts newlineto ensure data integrity, tamper-proof storage, and energy-efficient validation. The newlineML-LMP model predicts system failures proactively, outperforming traditional neural newlinearchitectures such as Multilayer Perceptron (MLP), Restricted Boltzmann Machine newline(RBM), Deep Belief Network (DBN), and Deep Neural Network (DNN) in accuracy, newlineconvergence speed, and robustness. Overall, this framework ensures timely and newlinereliable EHR access while maintaining secure, low-latency operations suitable for newlinereal-time clinical environments. newlineThe second framework presents a hybrid Convolutional Neural Network (CNN) newlineTransformer approach for detecting and predicting attack severity in EHMS. The newlineCNN module cap

Description

Keywords

Citation

item.page.endorsement

item.page.review

item.page.supplemented

item.page.referenced