The Relating Factor Analysis Of The Postmenopausal Women With Metabolic Syndrome And The Study Of The Early Predictive System

In recent years, with the increasing of aging population, metabolic syndrome has entered people's attention and become a research hotspot in today's society. Metabolic syndrome is not only a high risk factor of causing cardiovascular disease, but also one main reason for leading human death and disability. Especially in postmenopausal women, the phenomenon of the metabolic syndrome is much common. In this thesis, relevant factors in postmenopausal women with metabolic syndrome and the early prediction of diagnostic system are studied. The work of analyzing related factors in postmenopausal women with metabolic syndrome is mainly done by hospital expert. The main work in this paper is to complete the analysis of fat content on MRI image of thigh and design the prediction system of metabolic syndrome.Medical image segmentation algorithm is generally divided into two categories. One is based on regional information, and the other is based on boundary information. With the development of MRI technology in recent years, MRI can show soft tissues more and more clearly, which makes the accurate segmentation of fat possible. The thigh fat includes subcutaneous fat and intramuscular fat. Because of the connectivity and similar gray value between subcutaneous fat and parts of intramuscular fat, it's very difficult for most segmentation methods to separate the two in current time. Therefore, this article combines the expected maximum algorithm, fuzzy C means algorithm and level set algorithm to propose an automatic segmentation method based on improved level set algorithm on thigh MRI. The main purpose is to segment the subcutaneous fat tissue and intramuscular tissue of legs, which will provide physiological basis for exploring the cause of the metabolic syndrome. After verified by hospital experts, segmentation results can be used to study the metabolic syndrome.The predictive and diagnostic system for postmenopausal women with metabolic syndrome is designed in windows operating system environment, which uses C++ language and SQL Server 2000 database to develop. The system has functions to input, storage, search, modify and delete the patient information. The system can not only manage the information of record library but also predict and diagnose patients with metabolic syndrome at the same time. The results of system performance test show that the software interface is more attractive and easy to operate, and the system has good specificity and sensitivity in clinical data validation. Approved by the hospital experts, this system can be used in community and family after being improved.