基于T-Closeness的医疗档案数据脱敏与共享管控技术研究

Research on desensitization and sharing management and control technology of medical archives data based on T⁃Closeness

医疗数据会涉及到病人的隐私和医学伦理,针对目前医疗数据在网络共享时代所面临的数据隐私、数据所有权和使用权边界不清等问题,文中通过对医疗数据进行T-Closeness脱敏设计、对数据权限的边界进行分割,结合网络访问安全边界技术,改进了现有的网络安全架构,并提出一种增强医疗档案数据脱敏和共享管控的机制。仿真实验结果表明,所提方案在1500次并发访问的规模下,仍能较好地承受并发压力,脱敏任务的建立时长占总耗时约33%。相同参数条件下,服务器峰值CPU的利用率仅增加了约3%,服务器平均CPU利用率增加低于2%,服务器平均内存利用率增幅小于1%,增长均在可控范围内,且具有一定的工程应用价值。

Medical data will involve patient privacy and medical ethics.Aiming at the problems of data privacy,unclear boundary of data ownership and use right faced by medical data in the era of network sharing.This paper improves the existing network security architecture by T⁃Closeness desensitization design of medical data,segmentation of data authority boundary,combined with network access security boundary technology,and proposes a mechanism to enhance medical file data desensitization and sharing management and control.The simulation results show that the proposed scheme can still withstand the concurrent pressure under the scale of 1500 concurrent visits,and the establishment of desensitization task accounts for about 33%of the total time.Under the same parameter conditions,the peak CPU utilization of the server only increases by about 3%,the CPU utilization increases by less than 2%,and the average memory utilization of the server increases by less than 1%.The growth is within the controllable range,and has certain engineering application value.