融合联邦学习与区块链的医疗数据共享方案

doi:10.19678/j.issn.1000-3428.0061284

摘要/Abstract

摘要： 随着医疗大数据的发展，医疗数据安全、个人隐私保护等问题日益突出。为在高效利用各个医疗机构医疗数据的同时保护病人的隐私，提出一种将联邦学习与区块链相结合的医疗数据共享与隐私保护方案。使用联邦学习对多源医疗数据进行建模，将训练的模型参数和医疗机构的声誉值存储于区块链上，并利用区块链对贡献高质量数据的医院进行奖励。通过分析数据源质量对联邦学习算法性能的影响，提出一种基于双重主观逻辑模型的声誉值计算算法来改进联邦学习的精确度，使用改进的声誉机制保证在数据共享中筛选数据源的效率，并利用区块链和联邦学习技术，提高共享效率和实现隐私保护。此外，利用Tensorflow搭建分布式平台并对算法性能进行对比分析，实验结果表明，所提方案能够筛选出高质量的数据源，减少边缘节点与恶意节点的交互时间，即使当声誉值在0.5以上时，也能达到0.857的学习精确度。

关键词: 联邦学习, 声誉, 区块链, 双重主观逻辑模型, 医疗数据共享, 精确度

Abstract: With the development of medical big data, the security of medical data and personal privacy are becoming increasingly prominent.To efficiently use the medical data of various medical institutions and protect the privacy of patients, this study proposes a medical data-sharing and privacy-protection scheme that combines federated learning and blockchain.First, federated learning models the multi-source medical data and stores the trained model parameters and reputation value of the medical institution in the blockchain, which is used to support the high quality data of hospitals.Based on the analysis of the impact of data-source quality on the performance of the federated learning algorithm, a reputation value calculation algorithm based on a double subjective logic model is proposed to improve the accuracy of federated learning using an improved reputation mechanism to ensure the efficiency of screening data sources in data sharing, improve sharing efficiency, and achieve privacy protection through blockchain and federated learning.In addition, a distributed platform is built using Tensorflow for a comparative analysis of the algorithm performance.The results show that the proposed scheme can screen high-quality data sources, reduce the interaction time between edge and malicious nodes, and achieve a learning accuracy of 0.857, even when restricting the reputation value to be greater than 0.5.

Key words: federated learning, reputation, blockchain, dual subjective logic model, medial data sharing, accuracy

中图分类号:

TP391.41

温亚兰, 陈美娟. 融合联邦学习与区块链的医疗数据共享方案[J]. 计算机工程, 2022, 48(5): 145-153,161.

WEN Yalan, CHEN Meijuan. Medical Data Sharing Scheme Combined with Federal Learning and Blockchain[J]. Computer Engineering, 2022, 48(5): 145-153,161.

https://www.ecice06.com/CN/Y2022/V48/I5/145

图/表 7

20220723173934

20220723173938

20220723173943

20220723173947

20220723173950

20220723173953

20220723173956

参考文献

[1] ZHENG X C, MUKKAMALA R R, VATRAPU R, et al.Blockchain-based personal health data sharing system using cloud storage[C]//Proceedings of the 20th International Conference on E-Health Networking, Applications and Services.Washington D.C., USA:IEEE Press, 2018:1-6.
[2] LIU X J, XIA Y J, YANG W, et al.Secure and efficient querying over personal health records in cloud computing[J].Neurocomputing, 2018, 274(3):99-105.
[3] LIU X H, LIU Q, PENG T, et al.Dynamic access policy in cloud-based personal health record systems[J].Information Sciences, 2017, 379:62-81.
[4] LIU Z L, WENG J, LI J, et al.Cloud-based electronic health record system supporting fuzzy keyword search[J].Soft Computing, 2016, 20(8):3243-3255.
[5] LIU X G, WANG Z Q, JIN C H, et al.A blockchain-based medical data sharing and protection scheme[J].IEEE Access, 2019, 7:118943-118953.
[6] WALTERS R.Cyber attacks on U.S.companies in 2014[EB/OL].[2021-02-20].http://thf_media.s3.amazonaws.com/2014/pdf/IB4289.pdf.
[7] JIN H, LUO Y, LI P L, et al.A review of secure and privacy-preserving medical data sharing[J].IEEE Access, 2019, 7:61656-61669.
[8] JIN H, XU C, LUO Y, et al.Toward secure, privacy-preserving, and interoperable medical data sharing via blockchain[C]//Proceedings of the 25th International Conference on Parallel and Distributed Systems.Washington D.C., USA:IEEE Press, 2019:852-861.
[9] WANG Y, ZHANG A Q, ZHANG P Y, et al.Cloud-assisted EHR sharing with security and privacy preservation via consortium blockchain[J].IEEE Access, 2019, 7:136704-136719.
[10] 牛淑芬, 韩松, 于斐, 等.区块链电子病历中基于密钥聚合的密文检索方案[J].计算机工程, 2021, 47(5):36-43. NIU S F, HAN S, YU F, et al.Ciphertext retrieval scheme based on key aggregation for electronic medical record on blockchain[J].Computer Engineering, 2021, 47(5):36-43.(in Chinese)
[11] MCMAHAN H B, MOORE E, RAMAGE D, et al.Communication-efficient learning of deep networks from decentralized data[EB/OL].[2021-02-20].https://arxiv.org/abs/1602.05629.
[12] SILVA S, GUTMAN B A, ROMERO E, et al.Federated learning in distributed medical databases:meta-analysis of large-scale subcortical brain data[C]//Proceedings of the 16th International Symposium on Biomedical Imaging.Washington D.C., USA:IEEE Press, 2019:270-274.
[13] MARTINEZ I, FRANCIS S, HAFID A S.Record and reward federated learning contributions with blockchain[C]//Proceedings of International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery.Washington D.C., USA:IEEE Press, 2019:50-57.
[14] LU Y L, HUANG X H, DAI Y Y, et al.Blockchain and federated learning for privacy-preserved data sharing in industrial IoT[J].IEEE Transactions on Industrial Informatics, 2020, 16(6):4177-4186.
[15] HIEU N Q, ANH T T, LUONG N C, et al.Resource management for blockchain-enabled federated learning:a deep reinforcement learning approach[EB/OL].[2021-02-20].https://arxiv.org/abs/2004.04104.
[16] UR REHMAN M H, SALAH K, DAMIANI E, et al.Towards blockchain-based reputation-aware federated learning[C]//Proceedings of 2020 IEEE Conference on Computer Communications Workshops.Washington D.C., USA:IEEE Press, 2020:183-188.
[17] KUMAR R, KHAN A A, KUMAR J, et al.Blockchain-federated-learning and deep learning models for COVID-19 detection using CT imaging[J].IEEE Sensors Journal, 2021, 21(14):16301-16314.
[18] 闾海荣, 黄琼霞, 石顺中, 等.一种基于区块链和联邦学习的医疗数据安全共享方法:中国, 111698322A[P].2020-09-22. LÜ H R, HUANG Q X, SHI S Z, et al.Medical data security sharing method based on block chain and federated learning:CN111698322A[P].2020-09-22.(in Chinese)
[19] SHAYAN M, FUNG C, YOON C J M, et al.Biscotti:a ledger for private and secure peer-to-peer machine learning[EB/OL].[2021-02-20] https://www.researchgate.net/publication/329207243_Biscotti_A_Ledger_for_Private_and_Secure_Peer-to-Peer_Machine_Learning.
[20] LI T, SAHU A K, ZAHEER M, et al.Federated optimization in heterogeneous networks[EB/OL].[2020-02-20].https://arxiv.org/abs/1812.06127.
[21] NISHIO T, YONETANI R.Client selection for federated learning with heterogeneous resources in mobile edge[C]//Proceedings of International Conference on Communications.Washington D.C., USA:IEEE Press, 2019:1-7.
[22] LIU Y N, LI K Q, JIN Y W, et al.A novel reputation computation model based on subjective logic for mobile ad hoc networks[J].Future Generation Computer Systems, 2011, 27(5):547-554.
[23] KANG J W, XIONG Z H, NIYATO D, et al.Incentive mechanism for reliable federated learning:a joint optimization approach to combining reputation and contract theory[J].IEEE Internet of Things Journal, 2019, 6(6):10700-10714.

选择文件类型/文献管理软件名称

选择包含的内容