基于云计算的医疗诊断平台的设计与研究

发布时间：2018-03-22 03:08

本文选题：医疗诊断　切入点：云计算　出处：《天津工业大学》2016年硕士论文　论文类型：学位论文

【摘要】：随着生活水平的提高,人类对健康、医疗和日常保健质量的要求将越来越高,而传统的医疗诊断模式无法给患者提供全方位实时精准的健康检测。伴随着人口老龄化程度的加剧,生理数据量的规模将出现井喷式增长,传统的以医院为监控和管理中心的医疗诊断模式已经无法满足目前面向家庭和个人的实时监控需求。如何对大量的实时生理数据进行快速分析诊断,对医疗诊断平台的设计与研究提出了新的挑战。云计算的出现为有效地解决这些问题提供了希望。结合云计算强大的后台处理功能,海量存储空间等服务特征,本文在现有云计算技术的基础上提出构建在线医疗诊断平台的研究方案,包括平台的构建,云平台数据流程设计,以及生理数据实时分析性能的优化。平台的基础架构采用最新的Storm架构,很好的解决了大规模生理数据实时分析的问题,并利用HBase非关系型数据库构建海量的生理数据存储平台,以便后期对数据进行挖掘分析。在生理数据分析方面,本文对心电信号进行了详细的分析处理。针对心电信号中的基线漂移和工频干扰,提出零相位巴特沃斯滤波,很好地保证了信号无失真、无时域偏差。在QRS波的识别中,针对传统小波算法中R波错检漏检的问题,本文提出一种二次样条小波变换的方法,很好的解决了该问题。分析系统在云端直接调用Matlab进行并行计算分析,可有效利用现有的计算工具和计算资源,实现高性能的实时数据分析。整个系统的测试结果表明,在云计算的平台上进行生理数据分析可以有效地解决传统医疗诊断过程中服务模式独立、分散,诊断效率低,诊断结果不精确的问题,为将来实现医疗改革,构建新型医疗系统提供了具有参考价值的方案。
[Abstract]:With the improvement of living standards, human beings will have higher and higher demands on the quality of health, medical care and daily care. But the traditional medical diagnosis mode can not provide patients with comprehensive real-time and accurate health detection. With the aging of the population, the size of physiological data will increase by blowout. The traditional hospital monitoring and management center medical diagnosis mode can not meet the needs of real-time monitoring and control for families and individuals. How to analyze and diagnose a large number of real-time physiological data quickly. It presents a new challenge to the design and research of medical diagnostic platform. The emergence of cloud computing provides hope to solve these problems effectively. Combined with the powerful background processing function of cloud computing, massive storage space and other service features, Based on the existing cloud computing technology, this paper proposes a research scheme to build an online medical diagnosis platform, including the construction of the platform, the design of the data flow of the cloud platform, And the optimization of real-time analysis performance of physiological data. The platform infrastructure adopts the latest Storm architecture, which solves the problem of real-time analysis of large-scale physiological data. And using HBase non-relational database to build a massive physiological data storage platform, in order to mine and analyze the data in the later stage. In this paper, the ECG signal is analyzed and processed in detail. For baseline drift and power frequency interference in ECG signal, a zero-phase Butterworth filter is proposed, which can ensure the signal without distortion and time-domain deviation. In the recognition of QRS wave, a zero phase Butterworth filter is proposed. Aiming at the problem of R wave error detection in traditional wavelet algorithm, this paper presents a method of quadratic spline wavelet transform, which solves the problem well. The analysis system directly calls Matlab in the cloud for parallel computation and analysis. The high performance real-time data analysis can be realized by using the existing computing tools and computing resources effectively. The test results of the whole system show that, Physical data analysis on cloud computing platform can effectively solve the problems of independent, decentralized, low efficiency and inaccurate diagnostic results in the traditional medical diagnosis process, so as to realize the medical reform in the future. The construction of new medical system provides a plan with reference value.
【学位授予单位】：天津工业大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP393.09

【相似文献】