基于RAFT算法的数字医联体医疗信息跨平台共享方法

A cross-platform sharing method of digital medical conjoined medical information based on RAFT algorithm

随着数字医联体的快速发展,实现医联体内部各医疗平台间信息的安全共享是急需解决的问题之一。因此,研究基于RAFT算法的数字医联体医疗信息跨平台共享方法。在分析数字医联体组建架构基础上,构建集信息采集和管理、信息的整合与分配、平台服务、应用服务于一体的信息跨平台共享体系。其中,信息整合与分配过程采用RAFT算法通过节点选举与日志复制两个功能,实现医疗信息的跨平台过程中数据请求方和数据提供方间的数据分配和共享,并采用PageRank算法优化RAFT算法节点关键性,提高其信息共享的完整性。实验中,该方法的最大可容忍恶意节点数量可达到16个;在45s的时间内,该方法可以采集近700条信息数据;在节点数量为17个时,本文方法的通信复杂度仅为262。实验表明:该方法对恶意节点容忍量更高,对海量医疗信息的提取更快,跨平台共享信息过程的通信复杂度较低。

With the rapid development of the digital medical consortium,achieving secure sharing of information among various medical platforms within the consortium is one of the urgent problems to be solved.Therefore,a cross platform sharing method for medical information in the digital medical consortium based on the RAFT algorithm is studied.On the basis of analyzing the architecture of the digital medical consortium,build an information cross platform sharing system that integrates information collection and management,information integration and distribution,platform services,and application services.Among them,the information integration and allocation process uses the RAFT algorithm to achieve data allocation and sharing between data requesters and data providers in the cross platform process of medical information through two functions:node election and log replication.The PageRank algorithm is also used to optimize the node criticality of the RAFT algorithm and improve the integrity of its information sharing.In the experiment,the maximum tolerable number of malicious nodes in this method can reach 16;In 45 seconds,this method can collect nearly 700 pieces of information data;When the number of nodes is 17,the communication complexity of our method is only 262.The experiment shows that this method has a higher tolerance for malicious nodes,faster extraction of massive medical information,and lower communication complexity in the process of sharing information across platforms.