一种面向海量医疗数据范围查询的并行索引框架

A Partition-based Parallel Index Framework for Range-Query from Massive Medical Data

目的:随着近些年医院信息系统迅速推广使用,以及区域医疗、社保的兴起,大量数据中心、社保中心数据库建立,医疗数据量正以每日成千上万条记录的速度快速增长。同时,这些数据涉及大量与范围相关的查询操作,如地域、时间区段、生化指标区间等,如何有效满足对这类海量医疗信息的范围查询,成为各医疗机构及数据中心管理部门面临的亟待解决的问题。方法:提出一种基于数据分组的并行索引框架和算法。结果及结论:通过模拟实验发现该框架可以有效提高索引的批量更新性能及范围查询性能处理,为各种海量医疗信息范围查询应用场景的实现奠定了技术基础。

Objective： As more and more regional medical and social medical insurances are springing up and being integrated, massive heterogeneous medical data are collected and processed continuously through different channels in medical data center. The amount of data is increasing in thousands records daily. Meanwhile, dealing medical data involves many range-query requirements, such as region, time duration, biochemical analysis index interval, etc. The challenge is how to effectively meet the range-queW demand of massive medical information, which has become an urgent problem faced by the medical institutions and data center management. Methods： This paper presents a parallel indexing framework based on data partition. Result ＆ Conclusion： The experiment results show that this framework can give an efficient range-query of massive medical information and lay a technical foundation for all kinds of application scenarios.