Privacy Protection Scheme for Epidemiological Investigation Based on Blockchain Technology

Since Bitcoin was invented in 2008, the blockchain technology has been growing rapidly. A blockchain is a growing public ledger called blocks which are interconnected with a chain by utilizing cryptography. Utilizing blockchain, we can share information over the system safely, where everyone can see but can't do any alteration. Because of this feature of blockchain system, everyone can fetch or reason out the peer nodes' data or even private data. To protect the privacy and data security of the blockchain system, a common practice is to use privacy computing technology to guarantee the privacy protection required. At present, the global pandemic of the COVID-19 is still going on, and it will coexist with humanity for a long time. How to screen the flow regulation quickly and make it traceable is an important research topic. This paper proposed one epidemiological data privacy protection scheme based on blockchain platform combined with secure multi-party computation based on protocols and federated learning based on modern cryptography. In this paper, we use federated learning technology to realize the construction of a federated learning model involving multiple institutions and departments in the process of epidemiological investigation to avoid the risk of user privacy leakage during epidemiological investigation. The blockchain technology is used to realize the multi-institution federated learning model On-chain operation, with the introducing of the global model competition mechanism, to realize the optimal federated learning model On-chain, and ensure the optimization of learning goals. Meanwhile, the secret sharing technology is used to realize the sharing of the validation dataset, which avoids the optimization problem of federated learning model caused by the institution knowing the global validation dataset.