面向养老服务的电子健康档案系统设计与实现

发布时间：2018-06-23 11:48

本文选题：养老服务 + 电子健康档案　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：随着我国老龄化速度加快,养老服务需求不断增多,养老服务模式越来越多样化,而老人电子健康档案不够完善,缺乏数据规范,不能在多种养老模式之间共享老人的健康信息。因此,老人电子健康档案标准化管理变得尤为重要。为此本文将设计并实现面向养老服务的电子健康档案系统,将建立面向养老服务的电子健康档案数据模型,设计基于数据模型的数据融合与共享方案,提出基于Naive Bayes和Logistics Regression的老人健康分析模型的构建方法,详细设计关键功能模块并实现面向养老服务的电子健康档案系统。具体包括以下几方面工作:首先,针对面向养老服务的电子健康档案缺乏标准化管理的问题,提出面向养老服务的电子健康档案标准化需求,建立面向养老服务的电子健康档案数据模型,采集各个不同机构的健康数据项,抽象数据项的概念为类,标准化类的属性,通过值域代码表规范化属性的值域,从而指导数据库的建立和系统的实现。其次,针对老人健康数据在养老与医疗领域中缺乏互联互通的问题,在面向养老服务的电子健康档案数据模型的基础上,设计数据融合与共享方案,为系统设计公共Web API,并按照数据模型规定返回数据的格式,从而解决语义异构问题,并通过用户、菜单和角色的维护实现访问控制,防止对老人健康信息的非授权访问。再次,针对慢性病预测在养老领域中缺乏针对性的问题,提出老人健康分析模型,分别构建基于Naive Bayes的老人健康分析模型和基于Logistic Regression的老人健康分析模型,采用Logistic Regression方法提取老人特征属性,增加老人特征属性优化模型性能,使用Accuracy,Sensitivity和Specificity评价模型的综合性能,实验结果得出基于Naive Bayes的老人健康分析模型较好。最后,基于上述面向养老服务的电子健康档案数据模型,设计数据库,并实现面向养老服务的电子健康档案系统。采用基于数据模型的数据融合与共享方案实现共享门户网站功能;采用基于Naive Bayes的老人健康分析模型实现系统的老人健康分析功能;运用黑盒测试和白盒测试方法对系统的老人健康评估和分析等关键功能以及系统的性能进行测试。
[Abstract]:With the acceleration of aging in our country, the demand for old-age service is increasing, the mode of old-age service is becoming more and more diversified, and the electronic health file of the elderly is not perfect enough, lacking of data standard. Health information about the elderly cannot be shared between multiple pension models. Therefore, the standardized management of electronic health records for the elderly has become particularly important. Therefore, this paper will design and implement the electronic health file system for the aged, establish the data model of the electronic health file for the aged, and design the data fusion and sharing scheme based on the data model. Based on naive Bayes and Logistics Regression, an analysis model of elderly health is proposed. The key function modules are designed in detail and the electronic health file system oriented to pension service is implemented. The specific work includes the following aspects: firstly, aiming at the lack of standardized management of electronic health archives for pension services, the paper puts forward the need for standardization of electronic health archives for pension services. This paper establishes the data model of electronic health file for the aged service, collects the health data items of different organizations. The concept of abstract data items is the class, the attribute of the standardized class, and the range of the normalized attribute of the code list of the value range. In order to guide the establishment of the database and the implementation of the system. Secondly, aiming at the lack of interconnection between elderly health data in the field of old-age care and medical care, a scheme of data fusion and sharing is designed on the basis of the data model of electronic health archives for the aged. In order to solve the problem of semantic heterogeneity, the common Web API is designed for the system and the format of returning data is stipulated in the data model, and the access control is realized through the maintenance of users, menus and roles, and the unauthorized access to the health information of the elderly is prevented. Thirdly, aiming at the lack of pertinence in the field of chronic disease prediction in the field of old-age care, a health analysis model for the elderly is proposed, which is based on naive Bayes and Logistic Regression, respectively. The Logistic Regression method is used to extract the elderly feature attributes, to increase the performance of the model, and to evaluate the comprehensive performance of the models by using Accuracyence sensitivity and specificity. The experimental results show that the model based on naive Bayes is better for the elderly health analysis. Finally, the database is designed based on the above data model of the electronic health file for the aged service, and the electronic health file system for the aged service is implemented. The data fusion and sharing scheme based on data model is adopted to realize the function of shared portal, the elderly health analysis model based on naive Bayes is used to realize the function of elderly health analysis of the system. The key functions of the system, such as health evaluation and analysis, and the performance of the system, are tested by black box test and white box test.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：D669.6

【参考文献】