大数据技术在医院数据资源开发中的应用评价

通过问卷调查山东省内60家综合医院，统计分析医院数据开发与应用的基本情况、主要方向、平台建设、应用效果、经费投入与人才配置，阐述对比医院间差异，找出制约因素，提出改进意见，为相关医院深入拓展数据开发与大数据应用提供参考借鉴。     

2010年-2018年甲状腺结节相关SCI论文知识图谱及数据可视化分析

目的:分析2010年至2018年甲状腺结节相关SCI论文，并建模进行定性与定量分析。方法:利用文献计量学中的引文分析、共被引分析、数据可视化、聚类分析等方法，使用CiteSpace分析工具，分析来源于Web of ScienceTM核心合集数据库中有关论文的出版情况、国家、机构、作者、知识基础与研究热点。结果:到2018年10月31日截止，研究了4 618篇论文。发表论文数:美国在国家/地区中排名第一，延世大学在研究机构中排名第一，Kwak JY在所有作者中排名第一。从2010年开始，每年发表的论文数量都在稳步增长，有相当多的论文发表在如《新英格兰杂志》、《THYROID》、《JAMA》、《CELL》等高影响因子的期刊上。甲状腺结节的SCI论文参考文献聚类为6类，分别是papillary-like nuclear feature、current statu、radiofrequency ablation、ultrasound elastography、goiter area、data system。代表研究前沿的关键词是meta-analysis、recommendation、thyroid carcinoma、shear wave elastography与bethesda system。结论:本文通过对国际上甲状腺结节的研究基础、研究热点的分析，阐述了甲状腺结节的研究趋势，为我国医学工作者的研究提供参考。     

基于数据挖掘分析极光激酶A在食管癌中的表达及意义

目的:探讨极光激酶A(AURKA)在食管癌组织中的表达与功能,并分析其对患者生存预后的影响。方法:基于GEPIA及Oncomine数据库检索AURKA在食管癌组织与正常组织中的表达差异;通过STRING网络在线数据库构建有关AURKA的蛋白相互作用(PPI)网络,并将数据导入Cytoscape软件后采用Cyto Hubba分析插件筛选出核心基因;在GEPIA数据库中,基于TCGA和GTEx数据集分析AURKA与核心基因表达的相关性;采用DAVID网络分析工具对核心基因进行GO功能注释和KEGG通路富集分析;基于R2基因分析可视化平台分析AURKA的表达与食管癌患者生存预后的关系。结果:GEPIA、Oncomine数据库分析显示,AURKA在食管癌组织中的表达较正常组织明显升高(P<0.05),但与患者肿瘤分期无关(P>0.05);AURKA与CDC20 (r=0.78)、PLK1 (r=0.83)等核心基因的表达呈显著正相关,其生物学功能主要富集于细胞分化、DNA损伤检测点、细胞增殖、负性调控细胞凋亡等;参与细胞周期调控、细胞衰老、P53以及Fox O等信号通路的调节;AURKA高表达组与低表达组患者相比预后较差,差异具有统计学意义(P<0.05),且核心基因CCNB1、BUB1B、NEK2高表达组患者的生存率也显著低于低表达组(P均<0.05)。结论:AURKA在食管癌组织中表达升高,且高表达组患者预后不良;AURKA与CCNB1、NEK2等核心基因共同参与了肿瘤发生相关的功能与通路,AURKA有望成为食管癌患者早期诊断、治疗及判断预后的候选分子靶标。     

Measuring spatio-temporal accessibility to emergency medical services through big GPS data

Medical accessibility is an important indicator for evaluating the effectiveness of public health services. However, the previous medical accessibility studies mainly focus on spatial accessibility without considering temporal variation in population distribution which is significant for evaluating access to emergency medical service (EMS). This paper proposes a model of spatio-temporal accessibility to EMS called ST-E2SFCA based on adapting the enhanced two-step floating catchment area (E2SFCA) method. We apply our method to the greater Tokyo area for a large volume of GPS dataset with millions of users and compare the accessibility difference over space and time. To evaluate our model, we also analyze the distinction of our model over different weight sets and compare the performance of ST-E2SFCA with the traditional E2SFCA. The result shows that our method can illustrate the temporal difference and is suitable for measuring the spatio-temporal accessibility to EMS, thus can guide the hospital location selection and urban planning.     

过敏性紫癜性肾炎方药规律及中医证候分析

目的：通过对1993年1月至2019年12月CNKI、VIP、WanFang收录的中医治疗过敏性紫癜性肾炎的文献进行整理分析，探讨中医辨证论治本病的证候和方药规律。方法：采用SPSS 25.0统计软件，对文献中证候、方剂以及药物进行统计归类，总结各证候、方剂、药物出现的频次和所占比例，对药物和临床症状进行聚类分析。结果：共纳入文献93篇，通过整理分析得出39个中医证候，其中脾肾两虚、风盛血热、气不摄血、阴虚火旺、气阴两虚等为中医常见证候；证候要素分析：血热、风热、瘀血、湿热为常见实性病机；肾虚、脾虚、阴虚、气虚为常见虚性病机；病位主要在肾、脾、肺、肝四脏；共得方剂145首，其中自拟方53首，成方92首，二者以补益剂、清热剂、理血剂、祛风剂为主；中药206味，分为13类，以生地黄、牡丹皮、赤芍、当归、茯苓等为常用药。结论：过敏性紫癜性肾炎的病位在肾、脾、肺、肝；风、热、湿、毒、瘀、虚为本病的主要病机。在临床用药中以补虚药、活血药、清热解毒药、利水渗湿药为主，兼以应用祛风药；治疗中应采用中医辨证与辨病相结合的方式。     

一种面向中医药数据的高效脱敏算法＊

目的 针对中医药大数据平台传统的加密方案效率不高的问题，提出一种高效的基于属性的内积加密的数据脱敏算法。方法 Hash（哈希）算法是应用广泛的高效的数据加密方法，但传统的哈希算法基于单一的控制策略，效率不高。本文提出一种基于属性的内积加密的数据脱敏算法，把批量的敏感数据分割为不同长度数据颗粒度，与特定密文的哈希进行内积处理。结果 在面对中医药大数据平台的海量数据加密的场景，与传统的哈希加密算法相比，本文提供的加密算法具有很好的性能。结论 为了保障个人隐私数据不被泄露，中医药大数据平台中的个人医疗数据需要加密脱敏后，才能进行分析处理或对外发布。本文提出的算法具备灵活的数据颗粒度、策略和高效的性能表现，适用于海量的中医药数据脱敏。     

Large-scale analysis of interindividual variability in single and paired-pulse TMS data

ObjectiveThis study brought together over 60 transcranial magnetic stimulation (TMS) researchers to create the largest known sample of individual participant single and paired-pulse TMS data to date, enabling a more comprehensive evaluation of factors driving response variability.MethodsAuthors of previously published studies were contacted and asked to share deidentified individual TMS data. Mixed-effects regression investigated a range of individual and study level variables for their contribution to variability in response to single and paired-pulse TMS data.Results687 healthy participant’s data were pooled across 35 studies. Target muscle, pulse waveform, neuronavigation use, and TMS machine significantly predicted an individual’s single-pulse TMS amplitude. Baseline motor evoked potential amplitude, motor cortex hemisphere, and motor threshold (MT) significantly predicted short-interval intracortical inhibition response. Baseline motor evoked potential amplitude, test stimulus intensity, interstimulus interval, and MT significantly predicted intracortical facilitation response. Age, hemisphere, and TMS machine significantly predicted MT.ConclusionsThis large-scale analysis has identified a number of factors influencing participants’ responses to single and paired-pulse TMS. We provide specific recommendations to minimise interindividual variability in single and paired-pulse TMS data.SignificanceThis study has used large-scale analyses to give clarity to factors driving variance in TMS data. We hope that this ongoing collaborative approach will increase standardisation of methods and thus the utility of single and paired-pulse TMS.     

Pro-equity immunization and health systems strengthening strategies in select Gavi-supported countries

BackgroundAchieving universal immunization coverage and reaching every child with life-saving vaccines will require the implementation of pro-equity immunization strategies, especially in poorer countries. Gavi-supported countries continue to implement and report strategies that aim to address implementation challenges and improve equity. This paper summarizes the first mapping of these strategies from country reports.MethodsThirteen Gavi-supported countries were purposively selected with emphasis on Gavi’s priority countries. Following a scoping of different documents submitted to Gavi by countries, 47 Gavi Joint Appraisals (JAs) for the period 2016–2019 from the 13 selected countries were included in the mapping. We used a consolidated framework synthesized from 16 different equity and health systems frameworks, which incorporated UNICEF’s coverage and equity assessment approach – an adaptation of the Tanahashi model. Using search terms, the mapping was conducted using a combination of manual search and the MAXQDA qualitative analysis tool. Pro-equity strategies meeting the inclusion criteria were identified and compiled in an Excel database, and then populated on a tableau visualization dashboard.ResultsIn total, 258 pro-equity strategies were implemented by the 13 sampled Gavi-supported countries between 2016 and 2019. The framework determinants of social norms, utilization, and management and coordination accounted for more than three-quarters of all pro-equity strategies implemented in these countries. The median number of strategies reported per country was 17. Afghanistan, Nigeria, and Uganda reported the highest number of strategies that we considered as pro-equity.ConclusionFindings from this mapping can be useful in addressing equity gaps, reaching partially immunized, and ‘zero-dose’ vaccinated children, and valuable resource for countries planning to implement pro-equity strategies, especially as immunization stakeholders reimagine immunization delivery in light of COVID-19, and as Gavi finalizes its fifth organizational strategy. Future efforts should seek to identify pro-equity strategies being implemented across additional countries, and to assess the extent to which these strategies have improved immunization coverage and equity.