数据挖掘技术在冠心病早期预警系统中的应用研究

发布时间：2018-03-29 10:29

本文选题：冠心病　切入点：早期预警系统　出处：《河北大学》2017年硕士论文

【摘要】：随着经济社会的快速发展,在生活水平提高的同时生活节奏也大大加快,诊疗技术的发展却相对滞后,冠心病是目前威胁人类健康的主要疾病之一。但是往往由于人们工作的繁忙、医疗费用的昂贵、医生的缺少,使得很多人无法及时发现病情,贻误了最佳治疗时机。本论文是从具有多年临床经验的心内科医生所提供的、日常生活中能够容易获得的个人生理属性的大量数据中,运用数据挖掘技术得到生理属性的各参数间潜在的、有价值的规则,并且把这些规则应用到冠心病早期预警系统之中。该系统对于冠心病的早期预防和诊治具有重要意义。论文主要研究内容如下:1.从某三甲医院收集了大量冠心病患者病历和某高校学生家庭成员健康问卷调查表得到的健康人群的数据,并进行了数据整理,作为算法训练的样本。2.给出了一个基于BP神经网络的冠心病判别算法,目标就是通过测试者的各项属性值来判断其是否可能患有冠心病。首先,通过样本进行训练,设计网络模型结构,得到一个相对较好的神经网络模型。其次,根据生成的模型,计算出测试者是否有可能患有冠心病。3.使用朴素贝叶斯分析方法来预测患有冠心病的概率。分为两个步骤:第一步,计算各项属性不同取值的先验概率;第二步,根据测试者的输入信息,计算出患病概率。4.设计实现了冠心病预警原型系统。主要由两部分组成,第一部分是人机接口部分,用于输入预警系统所需要的个人身体状况的基础信息,并进行数据的完整性判断。第二部分是系统对冠心病的预测部分,根据输入的基本信息预测出冠心病的患病情况及健康建议。5.通过算法实验和软件系统测试,验证了原型系统的有效性。使用本系统,可以随时根据自身情况来评估患病的风险,可以让测试者保持警惕,积极调解自身状态,还可以为医疗机构的诊断提供有价值的参考。
[Abstract]:With the rapid development of economy and society, the pace of life has been greatly accelerated while the standard of living has been improved, while the development of diagnosis and treatment technology has lagged behind. Coronary heart disease (CHD) is one of the main diseases threatening human health at present. However, due to the busy work of people, the high cost of medical treatment and the lack of doctors, many people are unable to detect the disease in time. This paper is based on a wealth of data from cardiologists with many years of clinical experience who can easily obtain personal physiological properties in their daily lives. Using data mining technology to get potential and valuable rules between the parameters of physiological attributes, These rules are applied to the early warning system of coronary heart disease. This system is of great significance for the early prevention and treatment of coronary heart disease. The main contents of this paper are as follows: 1. A large number of coronary heart disease has been collected from a third class hospital. Patients' medical records and the data of healthy people obtained from a questionnaire on the health of family members of a college student, As a training sample of algorithm. 2. A BP neural network based coronary heart disease discrimination algorithm is presented. The goal is to determine whether the person is likely to have coronary heart disease through each attribute value of the tester. First of all, Through the training of samples, the network model structure is designed, and a relatively good neural network model is obtained. Secondly, according to the generated model, Use naive Bayesian analysis to predict the probability of coronary heart disease. There are two steps: the first step is to calculate the priori probability of different values of each attribute; the second step is to predict the probability of coronary heart disease. According to the input information of the tester, the probability of disease is calculated. 4. The prototype system of coronary heart disease warning is designed and implemented. The system is composed of two parts, the first part is the man-machine interface. It is used to input the basic information of the individual's physical condition needed by the early warning system and to judge the integrity of the data. The second part is the system's prediction of coronary heart disease. Based on the input of basic information to predict the prevalence of coronary heart disease and health advice .5.Through algorithm experiments and software system tests, the effectiveness of the prototype system is verified. Using this system, the risk of the disease can be assessed at any time according to their own conditions. It can keep the testers alert, actively mediate their own status, and can provide valuable reference for the diagnosis of medical institutions.
【学位授予单位】：河北大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R541.4;TP311.13

【参考文献】