Enhancing security and scalability in electronic medical records (emr) management: Integrating blockchain and machine learning

Abstract

The management of electronic medical records (EMRs) is a crucial yet challenging aspect of modern healthcare. In today's world, data security, privacy, and transparency are of utmost importance. This research proposes a blockchain-based knowledge system for EMR management, utilizing the Hyperledger Fabric framework and machine learning (ML) algorithms. By combining the immutable and  decentralized nature of blockchain with the predictive capabilities of ML, this system aims to support medical diagnostics and decision-making, with a specific focus on stroke prediction. Experimental results using over 130,000 real-world anonymized EMRs have demonstrated that this system significantly improves data sharing among healthcare facilities while maintaining security and data integrity. The system's decision tree model achieved an impressive 85% accuracy in predicting stroke risk, highlighting the potential for blockchain-ML integration to revolutionize healthcare management and diagnostics.c