基于二项Logistic回归建模的在线高血压病情分类系统的研究与开发

发布时间：2018-07-01 20:15

本文选题：在线医疗 + 分词　；参考：《内蒙古大学》2017年硕士论文

【摘要】：随着如今物质生活水平的不断提高,以及人们在饮食习惯上的改变,导致一些非传染性疾病对人类身体健康威胁越来越大,其中高血压对人类身体健康的威胁尤为严重。本文通过提取在线高血压患者的一些数据,然后挖掘、分析出其中潜在的隐藏信息,并对患者的患病程度进行分类。基于这些数据进行分类模型的构建,最后可以利用该模型对患者的患病程度进行预测。患者通过分类模型的预测结果能够对自身的患病情况有更准确的了解,从而选择合适的治疗方式。本文的研究目标是基于在线高血压患者的病情描述,以及医生给出的治疗方案来进行分类模型的构建,然后利用模型开发高血压在线病情自动诊断系统。该系统可将高血压患者的患病程度分为三类:0(无病)、1(轻微)、2(严重)。本文的研究过程由数据预处理、模型构建、系统开发三个模块组成。具体内容如下:(1)对文本数据进行分词处理。(2)提取患者在病情描述中的一些属性(如年龄、性别等),同时对医生给出的治疗建议进行分类处理。(3)进行数据转换,将文本数据转换为数值型。(4)两个模型的构建,模型一用来对患者无病或者有病进行分类,模型二用来确定有病患者的患病程度是轻微还是严重。(5)基于模型一、模型二进行高血压在线病情自动诊断系统开发。本文基于二项Logistic回归模型所开发的高血压在线病情自动诊断系统不仅可以对高血压患者的患病程度进行更准确的预测,同时对在线医疗服务体系也有着一定的理论和现实意义。
[Abstract]:With the improvement of material living standards and the change of people's eating habits, some non-communicable diseases are threatening human health more and more, especially hypertension. In this paper, we extract some data of patients with hypertension online, then mining, analyze the potential hidden information, and classify the degree of disease of patients. The classification model based on these data can be used to predict the degree of disease. By predicting the classification model, the patients can know their own disease more accurately and choose the appropriate treatment. The purpose of this paper is to construct the classification model based on the description of the condition of the patients with hypertension online and the treatment scheme given by the doctor, and then to develop the automatic diagnosis system of the condition of hypertension online by using the model. The system classifies hypertension patients into three groups: 0 (mild) and 2 (severe). The research process consists of three modules: data preprocessing, model building and system development. The specific contents are as follows: (1) partitioning the text data. (2) extracting some attributes (such as age, sex, etc.) of the patient in the description of the disease, and at the same time classifying the treatment advice given by the doctor. (3) data conversion. (4) the construction of two models. Model 1 is used to classify the patients who are not ill or sick. Model 2 is used to determine whether the degree of illness is mild or serious. (5) based on model 1, Model 2 was used to develop an online automatic diagnosis system for hypertension. The online automatic diagnosis system of hypertension based on binary Logistic regression model can not only predict the degree of hypertension more accurately, but also has certain theoretical and practical significance to the online medical service system.
【学位授予单位】：内蒙古大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R544.1;TP391.1

【参考文献】