Graves眼病相关因素的关联分析

 目的进行Graves眼病相关因素研究,通过数据挖掘形成关联规则。方法甲状腺疾病相关资料来源于中国医科大学
附属第一医院病案室,收集该院2000～2009 年的甲状腺疾病住院病例数据。内容包括与甲状腺疾病相关的15 项临床指标。首先
将各项指标数据离散化为计数资料,将源数据库映射为挖掘数据库;然后设置最小支持度为0.1,最小置信度为0.8,利用Apriori
算法进行关联规则挖掘,最后由临床免疫科专家解释及检验所产生的规则。结果共形成241 个强关联规则,这些强关联规则
中蕴含着Graves眼病发病与甲状腺功能系列及血常规指标、性别、年龄等因素之间的关联关系。结论通过数据挖掘方法挖掘
目前医院中大量医疗病例,从而发现其中蕴含的新信息、新知识,为医院大量病案信息提供了新的研究方法,具有一定的实用价
值。     

大数据及其对医学图书馆的启示

大数据时代的来临给传统图书馆带来了强烈冲击。国内外已有不少学者对大数据时代下的图书馆进行了研究,但医学专业图书馆如何应对大数据挑战的报道较少。在理性认识“大数据”概念和特点的基础上,分析了医学图书馆可以利用的大数据理论与技术,探讨了大数据给医学图书馆带来的机遇及需要解决的问题与对策,最后设计了大数据在医学图书馆的应用架构。     

基于数据挖掘技术分析腹泻型和便秘型肠易激惹综合征针灸使用规律

目的:通过文献分析临床针灸治疗腹泻型(D-IBS)和便秘型肠易激惹综合征(C-IBS)的经脉、腧穴选择特点和规律。方法:检索中国知网、万方、维普、Pubmed、Web of Science等数据库,收集2000—2016年针灸治疗D-IBS和C-IBS的临床研究报道,提取其所使用的针灸方式、经脉和腧穴。运用描述性分析、关联性分析和聚类分析研究其内在联系与规律规律,同时观察针灸治疗D-IBS和C-IBS的经脉、腧穴各自的规律和它们之间的差异。结果:共纳入针灸治疗D-IBS文献56篇、针灸治疗C-IBS文献6篇,D-IBS的治疗主要用针刺,临床以胃经和任脉的腧穴为主,如天枢、足三里、中脘,关联性分析发现胃经—任脉—脾经的为最常见的经脉组合,腧穴以足三里—天枢—太冲为主要关联群。聚类分析经脉主要分成2个核心聚类群,2个核心群组,腧穴则可以分成5个聚类群,5个核心群组。而C-IBS的研究则主要以悬灸为主,胃经、膀胱经的腧穴应用最多,如天枢、足三里、大肠俞、上巨虚,关联性分析发现经脉组合与D-IBS相同,也是胃经—任脉—脾经组合支持度和置信度最高,而腧穴则是以足三里—天枢的配伍最为常见。聚类分析则发现经脉、腧穴均可以分成2个核心聚类群,2个核心群组。结论:通过分析针灸治疗D-IBS和C-IBS的临床研究,认为虽然两者在针灸方法的使用上略微存在差异,D-IBS针灸并用,C-IBS以灸为主。但是在经脉和腧穴选择上却存在高度相似性。     

基于数据挖掘的支气管哮喘针刺组方选穴规律研究

目的:通过运用数据挖掘技术,探讨哮喘针刺选穴规律及核心配伍.方法:通过检索中国知网、万方数据库、维普数据库、中国生物医学文献数据库及PubMed数据库中1980年1月至2018年11月针刺治疗支气管哮喘的文献,利用Excel建立数据库,应用SPSS statistics、SPSS modeler软件对其进行频次分析、系...     

Schneeberg lung disease and uranium mining in the saxon ore mountains (Erzgebirge)

The so-called Schneeberg lung disease is a form of bronchial or alveolar carcinoma caused by the effects of the radioactive gas radon and of its radioactive short half-life daughter products. This type of radiation-induced occupational cancer is the most common and the most important radiation injury among workers occupationally exposed to ionizing radiation. There have been many deaths from lung cancer, especially in the Soviet uranium mines in the Erzgebirge of Saxony in the former German Democratic Republic. The history of disease in these miners extends over five centuries; the first observations of their health hazard start in the Middle Ages. The discovery of the lung cancer component was made toward the end of the nineteenth century, and the suspicion that a connection might exist between this cancer type and exposure to ionizing radiation was voiced at the beginning of the twentieth century. In the first half of this century, further research was carried out on this disease in the Schneeberg area of the Erzgebirge. Before the end of World War II, guidelines were set up to define the acceptable limits of radon exposure in the ore mines of Saxony. After World War II, the American uranium mines in the Colorado Plateau used the German research results as a basis for working out their own radiation protection standard. The uranium mines under Soviet occupation in the former GDR, on the contrary, paid no attention to these research findings. For many years, no precautions were taken for the miners' working conditions. The consequence of this serious omission was an estimated 9,000 fatal cases of lung cancer among these underground miners. High concentrations of radon are to be found in indoor air of homes in some districts of the Erzgebirge, suggesting an increasing lung cancer risk for the local inhabitants. The significance of this finding is evaluated. © 1993 Wiley-Liss, Inc.     

RadMonitor: Radiology Operations Data Mining in Real Time

This paper describes the web-based visualization interface of RadMonitor, a platform-independent web application designed to help manage the complexity of information flow within a health care enterprise. The system eavesdrops on Health Layer 7 traffic and parses statistical operational information into a database. The information is then presented to the user as a treemap—a graphical visualization scheme that simplifies the display of hierarchical information. While RadMonitor has been implemented for the purpose of analyzing radiology operations, its XML backend allows it to be reused for virtually any other hierarchical data set.     

Workflow mining for visualization and analysis of surgeries

Objective  Modeling the workflow of a surgery is a topic of growing interest. Workflow models can be used to analyze statistical properties of a surgery, for intuitive visualization, evaluation and other applications. In most cases, workflow models are created manually, which is a time consuming process that might suffer from a personal bias. In this work, an approach for automatic workflow mining is presented. Materials and methods  Ten process logs, each describing a single instance of a laparoscopic cholecystectomy, are used to build a Hidden Markov Model (HMM). Using a merging approach, models at different levels of detail are generated. These embody statistical information concerning aspects like duration of actions or tool usage during the surgery. Results  A Graphical User Interface (GUI) is presented, that uses a graph representation of the HMM to intuitively visualize surgical workflow. It allows changing the level of detail by expanding and merging nodes. The GUI can also be used to compare videos of surgeries which are synchronized to the model. Conclusions  The proposed method allows automatic generation and visualization of a statistical model describing the workflow of a surgery.     

Mining Associations Between Genetic Markers,Phenotypes, and Covariates

We used Haplotype Pattern Mining, HPM [Toivonen et al., Am J Hum Genet 67:133–45, 2000], for gene localization in Genetic Analysis Workshop (GAW) 12 isolate data. In HPM, association is analyzed by searching all trait‐associated haplotype patterns. Data mining algorithms are utilized to make the search efficient. The strength of the haplotype‐trait associations is measured by a linear model, into which a pre‐selected set of covariates is incorporated. Marker‐wise patterns of association are used for predicting the disease gene location. Genome‐wide scans of susceptibility genes for affection status as well as for the quantitative traits (Q1–Q5) were performed. First analyses were made with small sample sizes, 63–94 trios per trait, which is compared with a pilot study of a larger complex disease‐mapping project. Subsequently, the analysis was repeated with approximately 600 cases and 600 controls per trait to give higher power to the analyses. With small sample sizes, only the susceptibility genes having the strongest effects on the traits could be localized. The larger sample size gave very good results: all susceptibility genes, except one, could be correctly localized. First experiments on candidate genes suggested that HPM is applicable even to fine mapping of mutations in DNA sequence. © 2001 Wiley‐Liss, Inc.     

卷烟主流烟气及其物理化学检测数据的挖掘与信息抽提

应用因子分析与典型相关分析方法对卷烟生产企业批量样品实际检测的18个检测指标进行了数据挖掘与信息抽提。因子分析表明:可用12个独立的公共因子来描述卷烟质量,其中主流烟气中的总粒相物、烟碱、焦油与物理指标稀释率、闭式吸阻相关性很强,且可用第一主因子来体现其对卷烟质量的影响;第二主因子主要体现了总氮的信息,前两个主因子占总信息量的66%。典型相关分析表明:卷烟主流烟气与物理化学信息间存在显著的整体相关,其中反映主流烟气信息的U典型变量主要来自总粒相物、焦油、烟碱、水分、水分含率及抽吸口数的贡献;反映物理、化学指标信息的V典型变量主要来自稀释率、闭式吸阻、单支重及总糖、氯等化学指标的贡献,其中稀释率的作用最大。而CO、圆周与硬度对卷烟主流烟气与理化性质两类信息间的整体相关性的影响不大。有关结果对卷烟的降焦减害、建立卷烟主流烟气信息与理化信息间定量关系提供了重要信息和理论依据。