药物管理与处方点评系统的设计与实现

发布时间：2019-01-22 09:52

【摘要】：对医疗工作而言,处方的正确是很重要的,如有不慎甚至会造成致命后果,因而有必要对处方进行点评。但目前大多数医院的处方点评还在依靠人力,而未能利用信息技术。即使部分医院使用了的处方点评系统,也是功能固定,难以二次开发;或成本高昂,并非一般机构所愿意承担的。本文通过与调查分析,研究了在B/S架构下,开发新的处方点评系统的条件和可行性,并详细描述了我们已经开发且完成了主要功能的系统。我们开发的处方点评系统能够实现的功能包括:导入处方和药品数据;由管理者分配任务,让操作者进行手动点评之后审核。这部分的处方是自动随机分配的,因而减少了准备的花费。不仅如此,通过数据库中的药品信息和用药规则,系统能对大量处方进行自动点评。而自动点评的规则,又可以在线修改,或由文本程序进行修改。系统还提供平台化的服务,通过API,外来程序也可以通过我们的系统,来自动分析处方。此外,在我们的开发中,还有基于神经网络的对用药规则(用量为主)的参数生成的尝试。在结构方面,该系统基于LAMP架构,使用改进的MVC开发模式,并在数据库中用JSON存储复杂数据。前端使用HTML、Java Script及CSS,而服务端使用PHP语言。另外还有少数功能如处方及药品导入,是通过Java实现的。软件本身的体积并不大,属于轻量级的应用程序,可以用于小型、低端配置的设备。该系统兼具部分大型网站架构特性,有中转服务器,能分布处理,保证负载均衡。另外,可以确保各项事务的幂等性,一定程度上也具有原子性、一致性。该系统在界面设计方面也颇为考究,使用Bootstrap开发了符合标准的响应式界面,能在多数情况下确保操作的无误。在系统的安全方面,我们设计了防止SQL注入的机制。在功能测试时,我们使用了十余医生的数千条处方进行测试,包括对处方进行随机抽取、自动点评出具报告在内,功能均能正确实现。在文中,我们对该系统中所用的理论依据与开发技术,及其功能实现皆做了说明。对于数据库的结构、存储方式,各项技术的运用过程进行了详细的描述。另外,还对该系统未来可以做的事情做了简单的介绍。
[Abstract]:The correct prescription is very important to the medical work, such as carelessness or even fatal consequences, so it is necessary to comment on the prescription. But most hospitals still rely on manpower for prescription reviews rather than information technology. Even the prescription evaluation system used in some hospitals is fixed, difficult to develop again, or costly, and is not something that the average organization is willing to undertake. Through investigation and analysis, this paper studies the conditions and feasibility of developing a new prescription evaluation system under the framework of B / S, and describes in detail the system that we have developed and completed the main functions. The functions of the prescription evaluation system we developed include importing prescriptions and drug data, assigning tasks to the managers and allowing operators to conduct manual review and review. This part of the prescription is automatically randomly assigned, thus reducing the cost of preparation. Not only that, through the database of drug information and drug use rules, the system can automatically comment on a large number of prescriptions. The rules of automatic review can be modified online or modified by a text program. The system also provides platform-based services that automatically analyze prescriptions through API, foreign programs as well as our systems. In addition, in our development, there is a neural network based on the use of drug rules (dosage-based) parameter generation attempt. In terms of structure, the system is based on LAMP architecture, uses improved MVC development mode, and uses JSON to store complex data in database. The front end uses HTML,Java Script and CSS, and the server uses PHP language. In addition, a few functions, such as prescription and drug import, are implemented through Java. The software itself is not large and is a lightweight application that can be used in small, low-end devices. The system also has some large-scale website architecture features, has transit server, can distribute processing, ensure load balance. In addition, it can ensure idempotent and, to a certain extent, atomicity and consistency of transactions. The system is also very sophisticated in interface design, using Bootstrap to develop a standard responsive interface, in most cases to ensure that the operation is correct. In terms of system security, we designed a mechanism to prevent SQL injection. In the functional test, we use thousands of prescriptions of more than ten doctors to test, including random sampling of prescriptions, automatic evaluation report, the function can be realized correctly. In this paper, the theoretical basis, development technology, and function realization of the system are explained. The structure of the database, storage methods, the application of various technologies are described in detail. In addition, the system can be done in the future to do a simple introduction.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP311.52;R95

【参考文献】