Blockchain-based Secure Medical Data Management and Disease Prediction

Healthcare systems based on the Internet of Things have an increasing demand for health sensing technology. To manage the data collected and sampled by medical devices, traditional centralized data management will lead to attacks such as single point of failure, which pose a security threat. Aiming at the problems of low data trust and uncontrolled data sharing in telemedicine, we proposed blockchain-based secure medical data management and disease prediction. To securely manage healthcare data, we carefully designed three-tier architecture. Specifically, in the user sensor layer, medical sensors monitor the patient status in real-time. In the storage layer, to protect the privacy of patients, we stored their data in blocks and quantified the medical data by using information entropy technology. In addition, in the blockchain layer, we also used smart contracts for application, authorization, and access control of health data to eliminate privacy leaks caused by internal and external security risks. The information summary is recorded on the blockchain to ensure the integrity of backtracking and anti-repudiation. We designed an extensible machine learning algorithm to predict disease types using a disease prediction model algorithm based on transfer learning. Security analysis and numerical results showed that the proposed scheme can effectively manage the safety data of telemedicine and predict the patient's future condition.