Research On Electronic Medical Record Sharing Technology Based On Blockchain

Due to the unbalanced distribution of medical resources,the medical level of large hospitals in developed areas is stronger than that of hospitals in underdeveloped areas,so sharing electronic medical records among different hospitals is an effective way to improve the level of medical services.In the existing studies using blockchain and attribute-based encryption algorithms to share electronic medical records and control the fine-grained sharing of electronic medical records,since attribute-based encryption algorithm requires trusted third parties to generate keys,trusted third parties have the ability to decrypting all the electronic medical records which cannot guarantee the confidentiality of the electronic medical records;at the same time,the anonymity of the blockchain makes the authenticity of the electronic medical records which are not guaranteed;in addition,the use of smart contracts to search the electronic medical records stored on the blockchain when the electronic medical record key words may reveal the electronic medical record information.This thesis studies these problems,and the main work and contributions are as follows.(1)An electronic medical record sharing scheme based on blockchain and ciphertext strategy-based attribute encryption(CP-ABE)is proposed in which the hospital acts as a key generator.Aim at the problem that third-party entities can decrypt all ciphertexts in attribute encryption,the hospital itself acts as the key generator.At the same time,in view of the fact that the hospital cannot be responsible for the electronic medical record due to the anonymity of blockchain,this thesis proposes a set of smart contracts for a hospital,and data users can search by using the smart contracts of the hospital,so that the users can know which hospital obtains the electronic medical record from,and the hospital that needs to be responsible for the electronic medical record.For the storage of electronic medical records,hierarchical storage structure is adopted.Electronic medical records are symmetrically encrypted and stored on an interplanetary file system(IPFS),while metadata is encrypted and stored on a blockchain by CP-ABE.According to the proposed scheme,an improved PBFT consensus algorithm based on Raft(RPBFT)is proposed,which is more suitable for the proposed scheme to reduce the communication consumption of consensus and increase the throughput.(2)An electronic medical record search smart contract based on Bloom filter is proposed.Because keywords may leak the information of electronic medical record during the searching process,it is proposed to use Bloom filter instead of keywords to search.Since the Bloom filter is a string of meaningless 0-1 arrays,it does not leak keyword information.A smart contract for accurate search of electronic medical records based on Bloom filter is proposed,which matches the keyword Bloom filter of electronic medical records and the electronic medical record keyword Bloom filter desired by data users.If the keywords are the same,the search results are returned.Aiming at the problem that the search fails due to the different number of keywords and the learning is meaningful even if only some of the keywords are the same in the precise search,a smart contract for fuzzy search of electronic medical records based on Bloom filter is proposed.(3)The experimental results show that the electronic medical record sharing model based on blockchain and CP-ABE proposed in this thesis can realize data sharing;At the same time,the model can ensure data privacy and security;the proposed RPBFT consensus algorithm is more efficient than the traditional PBFT consensus algorithm consumes which is less communication and has greater throughput.