基于SEER数据库利用机器学习方法分析乳腺癌的预后因素

目的以SEER数据库中1990—2014年间的乳腺癌数据为研究对象,利用机器学习方法,分析乳腺癌的预后因素,辅助医师对患者的预后进行有效评判。方法根据临床医师的建议,筛选了12个字段作为模型输入字段,以术后5年生存状况作为模型输出字段。首先利用单因素统计分析方法初步筛选预后因素,再分别利用logistic回归和决策树两种机器学习分类算法进行建模分析,藉此寻找影响乳腺癌5年预后的因素。采用十折交叉法组织样本数据,并利用过抽样和欠抽样技术进行样本的平衡处理;以灵敏度、特异度及ROC下的AUC等参数作为模型的评价指标。结果在12个模型输入字段中,肿瘤分期、肿瘤分级、肿瘤尺寸、雌激素水平、年龄分组、孕激素水平等因素对于乳腺肿瘤预后具有较大影响;在此两种模型下,模型测试集上的灵敏度和特异度均介于74.2%~78.2%之间,AUC均处于0.838~0.850之间。结论利用Logistic回归和决策树算法构建乳腺癌患者的优化预后模型,可辅助医师判断患者预后情况及治疗效果。     

四川省常用药品使用及储备情况入户调查

目的：从需求侧的角度分析新医改多重政策实施背景下四川省家庭药品可及性改善情况和抗菌药物的使用现状。方法：参考世界卫生组织推出的《药品可及性及药品使用情况入户调查指南》,在四川省抽取成都、绵阳、内江、南充、广安、巴中6个市,在每个市的2个城区、2个县城以及2个乡镇共抽取1 080户家庭为样本进行问卷调查,调查四川省常用药品使用及储备情况,并对收集的数据进行统计分析。结果：总体上,本次研究实际收集到1 103份有效问卷,458户受访家庭最近2周有成员患有急性病,521户受访家庭有成员患有慢性病。90%以上的急慢性病患者获得了药物治疗,受访家庭月均药品费用支出为334.33元,占每月家庭总支出的9.3%,不足20%的急慢性病患者获得了药品医保报销。40.1%的急性病患者使用了抗生素,并且47.5%使用的抗生素来自公立医疗机构。79.7%的受访家庭家里备有药品,且乡镇家庭储备药品的比例低于城区、县城家庭。结论：新医改多重政策实施背景下四川省常用药品可及性较好,急慢性病药品可负担性情况良好,但在医保报销水平和家庭抗生素合理使用方面需要不断完善。     

A searchable database for proteomes of oral microorganisms

An online database of proteomes for two-dimensional electrophoresis (2DE) gel data was constructed and it is now freely accessible through a web-based interface. Proteins from three oral bacteria, Streptococcus mutans UA159, Actinobacillus actinomycetemcomitans HK1651, and Porphyromonas gingivalis W83, whose genome databases are freely available, were separated by 2DE, and protein spots were analyzed by matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) and identified. About 1000 spots from the gels of P. gingivalis W83 were extracted and analyzed by MALDI-TOF, and 330 proteins were identified. In addition, 160 of 240 spots of A. actinomycetemcomitans and 158 of 356 spots of S. mutans were identified. Information such as spot coordinates on the gels, protein names (predicted functions), molecular weights, isoelectroric points, and links to online databases, including Oral Pathogen Sequence Databases of the Los Alamos National Laboratory Bioscience Division (ORALGEN) and National Center for Biotechnology Information (NCBI) or The Institute Genomic Research (TIGR), were stored in tables accessible through the relational database management system MySQL on an Apache web server. To test for functionality of this database system, responses of S. mutans to environmental changes were analyzed using the database and 21 spots on the gel were identified as proteins whose expression had been increased or decreased by environmental pH change without in-gel trypsin digestion, protein extraction, or MALDI-TOF/TOF-MS (mass spectrometer) analysis. The identified proteins are agreement with those reported in previous papers on acid tolerance of S. mutans, demonstrating the usefulness of the system. This database is available at http://www.myamagu.dent.kyushu-u.ac.jp/~bioinformatics/index.html or http://www.bipos.mascat.nihon-u.ac.jp/index.html.     

耳廓三维图像数据库的应用研究

目的：探索一种新的耳腰复体制作方法。方法：将耳廓三维图像数据库中的三维模型，在surfacer 10．0软件平台上数据处理为CAD模型；并在此基础上，根据预修改内容设计标准，局部修改调整耳廓三维图像数据库中三维模型形态，建立预修改模型库和一种模拟的CAD功能。在Magics RP5．41软件平台上完成耳廓三维模型按一定比例放大、缩小及镜像处理。运用快速成型技术完成最终三维模型向实体模型的转化过程，制作耳腰复体模型。结果：建立了一个耳廓三维图像数据库应用系统。结论：探索了一种新的，更加方便的制作耳腰复体的方法。     

基于医院信息系统和Oracle9i数据库的网络化口腔电子病历安全体系的研究

目的 探讨网络化口腔电子病历安全体系的实现方式. 方法在Windows 2000/XP平台上,结合基于HL7标准的XML语言、Oracle9i数据库构建多层分布式的网络化口腔电子病历系统,通过用户名和密码MD5技术加密、角色分配、分级权限、痕迹保留、审计追踪、DES数据加密、数据备份、实时监控、网络管理、服务器双机热备份等方法构建网络化口腔电子病历系统的安全体系.结果 本研究构建了基于局域网的网络化口腔电子病历系统安全体系,可确保电子病历信息录入、修改、存储和传输的安全. 结论该网络化口腔专科电子病历符合口腔专科医院信息系统安全管理的需要,有助于推动口腔专科医院的信息化管理建设.     

白蛋白微球颈外动脉灌注后经动—静脉短路溢出量及主要脏器分布的分析

目的　了解平均粒径 53 6μm的白蛋白微球 (AMS)经颈外动脉灌注后 ,经动 -静脉短路 (AVA)溢出部分微球的量和在重要脏器的分布。方法　经家兔颈外动脉灌注1 2 5I标记的微球后抽取颈外静脉血 ,测定放射剂量和静脉血中所含微球的比例 ;观察微球经家兔颈外静脉推注后全身各重要脏器组织中微球的分布情况。结果　经颈外动脉注入微球后仅有 0 51 %的放射剂量出现在静脉血中。微球经静脉推注后 ,主要聚集于肺 ,其它各重要脏器未见分布。结论　以平均粒径 53 6μm的白蛋白微球作为口腔颌面部肿瘤栓塞化疗药物的载体 ,全身副作用轻微 ,是安全可行的     

Access to dental care by young South Australian adults

BACKGROUND: Despite reported concern over the dental care of young adults little research has been done on their use of dental services in Australia. The aim of this study was to investigate the patterns of dental utilization of young South Australian adults aged 20-24 years. METHODS: A random sample of 2300 young adults was selected from the electoral roll. Partial or complete addresses and possible phone numbers were obtained for 1921 persons. Telephone interviews were conducted for 1261 subjects to obtain information on socio-demographics, health behaviour and dental visiting (response rate 65.6 per cent). RESULTS: One third of young adults (34 per cent) had not made a dental visit in the previous two years and 38 per cent usually visited for a problem rather than a check-up. Making a dental visit in the last two years was significantly associated with a number of socio-demographic variables including age and gender, with holders of private dental insurance and those who have not avoided care because of cost having higher odds of making a visit and males and government concession card holders having lower odds of visiting. Usual reason for visiting a dentist for a problem was significantly associated with no private dental insurance, holding a government concession card, no tertiary education and avoiding care because of cost. CONCLUSIONS: This study suggests that demographic and economic factors influenced use of dental services and reason for visiting of young South Australian adults.     

Global epidemiology, risk factors and prevention of oral cancer

The Globacan [2002] database from the International Agency for Research on Cancer, shows 400,318 cases of oral and pharyngeal [excluding nasopharynx] cancer in the world annually, and 221,917 deaths.     

关注miRNA及相关肿瘤数据库在肿瘤研究中的应用

肿瘤是一种严重威胁人类健康的疾病,其发生发展历经漫长而复杂的过程。深入探究肿瘤分子发病机制将有助于肿瘤早期诊断和开发新的抗肿瘤药物。小核糖核酸(miRNA)是一种小片段的RNA,它通过抑制靶基因表达影响肿瘤的发展进程。本文以miRNA为中心,系统阐述miRNA参与肿瘤进程的分子机制,miRNA在肿瘤治疗以及循环miRNA在肿瘤早期诊断和预后评估方面的应用前景;数据库在肿瘤相关分子机制研究及筛选肿瘤发展进程中miRNA调控关键基因的应用。希望通过本文与广大研究工作者共同努力推动肿瘤的机制研究。     

基于新型冠状病毒肺炎的临床数据治理和科研平台建设实践

我院作为新型冠状病毒肺炎患者收治的定点医院,积累了大量新型冠状病毒肺炎患者的医疗数据。为有效利用这些数据为临床关于新型冠状病毒肺炎科研攻关服务,我院自主研发基于新型冠状病毒肺炎的科研数据平台。利用ETL工具KETTLE整合新型冠状病毒肺炎患者的门诊和住院数据,跨平台采集和清洗HIS、电子病历、LIS和PACS等业务系统的数据,形成新型冠状病毒肺炎专病库。以多种数据关联查询以及统计分析为设计思路的科研平台,为临床医生探索性研究提供思路以及数据支撑,进一步提高了临床关于新型冠状病毒肺炎科研产出的速度和质量。