本文選題：醫(yī)療健康 + Android�。� 參考：《廣西師范大學(xué)》2017年碩士論文

【摘要】：隨著我國移動互聯(lián)網(wǎng)、物聯(lián)網(wǎng)、云計算、可穿戴設(shè)備等新技術(shù)的發(fā)展,以及惠及全民的健康信息服務(wù)和智慧醫(yī)療服務(wù)的提出,推動了健康大數(shù)據(jù)的應(yīng)用。面對海量的醫(yī)療數(shù)據(jù),高效地存儲和快速地處理數(shù)據(jù)成為當(dāng)前主要需求。傳統(tǒng)的數(shù)據(jù)庫對海量級的醫(yī)療數(shù)據(jù)面臨著存儲應(yīng)接不暇、成本居高不下、計算能力無法企及等問題。針對這些問題,本文利用Hadoop架構(gòu)實現(xiàn)對海量數(shù)據(jù)的分布式存儲和處理,結(jié)合集群特性實現(xiàn)低成本、高效、可擴展的數(shù)據(jù)處理。本文結(jié)合醫(yī)療健康數(shù)據(jù)的特性,提出基于Hadoop的醫(yī)療健康數(shù)據(jù)管理系統(tǒng),系統(tǒng)分為兩大模塊:基于Android的醫(yī)療健康管理系統(tǒng)客戶端和基于Hadoop的醫(yī)療健康數(shù)據(jù)管理系統(tǒng)服務(wù)器端。基于Android的醫(yī)療健康管理系統(tǒng)客戶端,是以Android智能手機為依托,設(shè)計移動互聯(lián)的健康管理APP。該軟件包括用戶注冊登錄、心率檢測、體重測量、運動測量、數(shù)據(jù)統(tǒng)計,健康建議六大部分,能實現(xiàn)生理參數(shù)的測量,改進運動測量算法,分析測量結(jié)果和提供健康建議等功能�；贖adoop的醫(yī)療健康數(shù)據(jù)管理系統(tǒng)服務(wù)器端,利用Hadoop全新技術(shù)對數(shù)據(jù)進行快速存儲和管理。本文搭建4臺機器的集群中心,考慮到醫(yī)療數(shù)據(jù)需要頻繁實時寫入讀取的特性,在Hadoop集群中心搭建HBase數(shù)據(jù)庫,取代HDFS,并重新對數(shù)據(jù)格式進行設(shè)計,從而實現(xiàn)數(shù)據(jù)庫HBase滿足存儲醫(yī)療數(shù)據(jù)。同時,使用并行計算模型MapReduce對醫(yī)療健康數(shù)據(jù)進行分析處理。此外,改進HBase多條件查詢方法,將HBase與Solr結(jié)合實現(xiàn)多條件查詢,提高查詢效率。最后,設(shè)計并模擬對醫(yī)療健康數(shù)據(jù)管理系統(tǒng)的客戶端和服務(wù)端的系統(tǒng)性能和可靠性測試,驗證結(jié)果表明系統(tǒng)能夠達到需求的功能設(shè)計和可靠性能,Android各個功能模塊測試通過,MapReduce性能測試和HBase數(shù)據(jù)庫寫入測試與預(yù)期一致,HBase和Solr相結(jié)合查詢數(shù)據(jù)時間大大縮短。因此,基于Hadoop的醫(yī)療健康數(shù)據(jù)管理系統(tǒng)能夠更好的滿足醫(yī)療健康數(shù)據(jù)要求。
[Abstract]:With the development of mobile Internet, Internet of things, cloud computing, wearable devices and other new technologies in China, as well as health information services and intelligent medical services that benefit the whole people, the application of health big data has been promoted. In the face of massive medical data, efficient storage and fast processing of data has become the main demand. The traditional database is facing the problems of overflowing storage, high cost and unreachable computing power for massive medical data. To solve these problems, this paper uses Hadoop architecture to realize the distributed storage and processing of massive data, and combines the characteristics of cluster to achieve low-cost, efficient and scalable data processing. Based on the characteristics of medical and health data, a medical and health data management system based on Hadoop is proposed in this paper. The system is divided into two modules: the client of medical and health management system based on Android and the server of health data management system based on Hadoop. The client of the medical health management system based on Android is based on the Android smart phone to design the mobile connected health management app. The software includes user registration, heart rate detection, body weight measurement, exercise measurement, data statistics, health advice six parts, can achieve the measurement of physiological parameters, improve the motion measurement algorithm, Analyze the results and provide health advice. The medical and health data management system server based on Hadoop uses the new technology of Hadoop to store and manage the data quickly. In this paper, the cluster center of 4 machines is built. Considering the frequent and real-time writing and reading of medical data, the HBase database is built in the Hadoop cluster center to replace the HDFS, and the data format is redesigned. In order to achieve the database HBase to meet the storage of medical data. At the same time, the parallel computing model MapReduce is used to analyze and process the health data. In addition, we improve the method of HBase multi-condition query, combine HBase and Solr to realize multi-condition query, and improve the efficiency of query. Finally, design and simulate the system performance and reliability test of the client and server of the medical and health data management system. The verification results show that the system can meet the requirements of functional design and reliability. The time of querying data through MapReduce performance test and HBase database writing test is greatly shortened with the combination of HBase and Solr. Therefore, the medical and health data management system based on Hadoop can better meet the requirements of medical and health data.
【學(xué)位授予單位】：廣西師范大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TP311.13;TP311.52

【參考文獻】

相關(guān)期刊論文 前10條

1 高榮偉;;“健康醫(yī)療大數(shù)據(jù)”正向我們走來[J];信息化建設(shè);2016年08期

2 龔世文;朱忠敏;;Android平臺下的空氣質(zhì)量和天氣情況查詢應(yīng)用設(shè)計與實現(xiàn)[J];電腦知識與技術(shù);2016年10期

3 支丹萍;;大數(shù)據(jù)對于企業(yè)的意義[J];通訊世界;2015年19期

4 關(guān)國棟;滕飛;楊燕;;基于心跳超時機制的Hadoop實時容錯技術(shù)[J];計算機應(yīng)用;2015年10期

5 周作建;林文敏;王斌斌;潘金貴;;基于海量醫(yī)療數(shù)據(jù)的癥狀自查服務(wù)云框架設(shè)計[J];計算機科學(xué)與探索;2015年09期

6 林碧英;王艷萍;;基于Hadoop的電力地理信息系統(tǒng)數(shù)據(jù)管理[J];計算機應(yīng)用;2014年10期

7 杜新星;;體育類大學(xué)生體成分及體質(zhì)健康狀況的調(diào)查分析[J];當(dāng)代體育科技;2014年15期

8 黃浩;;大健康產(chǎn)業(yè)的數(shù)據(jù)入口[J];中國信息化;2014年01期

9 姜浩端;;大數(shù)據(jù)的本質(zhì)及其可能的影響[J];中國經(jīng)濟報告;2013年06期

10 向方明;朱遵義;許敬;崔業(yè)兵;;YUV到RGB顏色空間轉(zhuǎn)換算法研究[J];現(xiàn)代電子技術(shù);2012年22期

相關(guān)碩士學(xué)位論文 前6條

1 戴，

本文編號：1932492