Research Of Clinical Data Mining And Analysis Based On Thyroid Disease

With the development of medical information technology, the hospital has accumulated a huge amount of clinical data resources in the process of providing medical services for patients. As more and more clinical data are stored in the database, how to mine the valuable information from these large amount of historical clinical data is a hot issue of the research in big data era. In order to solve this problem, scholars in China and abroad have applied the data mining technology to medical field in recent years.In this thesis, we analyze the current situation of the information system in a large tertiary hospital in Shanghai, and design the clinical data analysis platform, which includes the logical structure and physical structure design. The data mining technology is used to analyze the clinical data of thyroid disease based on this platform. The main contributions include the following four aspects:Firstly, according to the current situation of the hospital information system, the data of the existing business system is analyzed, and the clinical data analysis platform based on data warehouse is designed. Data analysis platform standardize data interface and integrate all medical data from clinical information system in the hospital. It can also lay a solid foundation for clinical data mining and analysis,assisting medical diagnosis and decision-making of different diseases by using computer.Secondly, the clinical data of patients were collected from the clinical medical data platform, including the basic information of the patients, the examination index data and the prescription. ETL technology, data cleaning, transformation, integration, etc., are used to pre process the raw data, and the data are analyzed and visualized after processing.Thirdly, in clinical medical diagnosis of thyroid diseases, the accurate diagnosis of disease is the key to cure the disease.This paper presents a classification method based on random forest using the clinical data of thyroid disease. The method of principal component analysis is used to select the data set and reduce the data dimension, then we use the random forest algorithm to complete the classification task. The method is characterized by the introduction of principal component analysis to reduce the dimension of thyroid disease data, which makes up for the shortcomings of the random forest algorithm in attribute selection.Finally, because of the correlation between different diseases, a type of disease often occurs with other diseases at the same time. In the course of treatment, it will also be used in combination with a variety of drugs. In order to solve these problems, this paper uses association rules algorithm for thyroid disease prescription drug and complications. The association rules provides a reference for the clinical treatment of drug selection and disease prevention, the results can reduce the cost of treatment and improve the treatment effect as well.