Machine learning for effective EHR management in blockchain-cloud integration

Abstract

<p>Machine learning (ML) techniques have gained prominence in effectively managing Electronic Health Record (EHR) systems within the context of blockchain-cloud integration. This study presents a hybrid Machine Learning approach that combines logistic regression (LR) and random forest (RF) techniques for EHR management, leveraging the data stored in a blockchain-cloud integrated system. The tamper-resistant nature of blockchain ensures the authenticity and security of the stored patient information, serving as a reliable source for learning. The proposed LR+RF model is evaluated against other algorithms, considering various performance metrics. The analysis reveals that the LR+RF model achieves an impressive accuracy rate of 98.37%, indicating its efficacy in accurately classifying EHR data and facilitating effective management. Furthermore, the study compares the performance of blockchain-cloud-based decentralized storage with blockchain-based storage and peer-to-peer storage in terms of latency and throughput. The results demonstrate that the blockchain-cloud integrated decentralized storage surpasses other storage methods, achieving an average throughput of 6.8 units and a latency of 4.7 units. These findings highlight the potential of the proposed LR+RF model for EHR management within a blockchain-cloud integrated environment. The use of blockchain as a secure storage environment ensures the integrity of patient information, while Machine Learning techniques enhance the accuracy of classification.</p>