Tacrolimus Inhibits Vasoconstriction by Increasing Ca^(2+) Sparks in Rat Aorta

The present study attempted to test a novel hypothesis that Ca^2＋ sparks play an important role in arterial relaxation induced by tacrolimus. Recorded with confocal laser scanning microscopy, tacrolimus（10 μmol/L） increased the frequency of Ca^2＋ sparks, which could be reversed by ryanodine（10 μmol/L）. Electrophysiological experiments revealed that tacrolimus（10 μmol/L） increased the large-conductance Ca^2＋-activated K＋ currents（BKCa） in rat aortic vascular smooth muscle cells（AVSMCs）, which could be blocked by ryanodine（10 μmol/L）. Furthermore, tacrolimus（10 and 50 μmol/L） reduced the contractile force induced by norepinephrine（NE） or KCl in aortic vascular smooth muscle in a concentration-dependent manner, which could be also significantly attenuated by iberiotoxin（100 nmol/L） and ryanodine（10 μmol/L） respectively. In conclusion, tacrolimus could indirectly activate BKCa currents by increasing Ca^2＋ sparks released from ryanodine receptors, which inhibited the NE- or KCl-induced contraction in rat aorta.

The Nobel Literature Prize Sparks a Mo Yan Craze

正At 19 o'clock, October 11th, Beijing time, the Swedish Academy awarded the 2012 Nobel Prize for Literature to the Chinese writer Mo Yan. Within several hours, Mo Yan's works were struck by the "Nobel effect", sales shooting up in both bookstores and online marts. Some websites also jumped on the bandwagon and put up a reservation service for the new editions.

Watch the Sparks Fly in Yuxian!

Hinterland rural towns are the last sanctuaries of old traditions, best honored at the Chinese New Year.YUXIAN in northwestern Hebei Province is a three-hour drive from Beijing.

Dog Meat Festival Sparks Controversy

A controversial annual dog meat festival is celebrated on summer solstice, falling on June 21 or 22, in Yulin, a city in south China＇s Guangxi Zhuang Autonomous Region. However, the festival has drawn widespread criticism in recent years.

Experimental Study on Methane Explosion Ignited by Sparks of Cable Bolt Breakage

An experimental device was designed for studying methane explosion ignited by sparks of cable bolt breakage. With the methane concentration being in explosion range, a series of experiments were conducted to study the law of spark generation during cable bolt breakage and the probability of methane explosion caused by the spark. The results show that the probability of generating sparks during cable bolt breakage is 50%. The spark generated by the breakage of steel cable bolt strand can't ignite a methane explosion. A detection was carried out using infrared-ray imaging apparatus (IRIA) to measure temperature of the spark generated by cable bolt breakage. It is indicated that the maximum temperature of the spark generated by cable bolt breakage is far less than the required ignition temperature for a methane explosion.

The Optical Diagnosis of Underwater Positive Sparks and Corona Discharges

In this paper, two types of underwater discharges, spark discharge and corona discharge, are investigated by optical diagnosis using a high speed framing camera （HSFC） with the framing time within nanoseconds under the same experimental conditions. In order to capture the photographs of streamer propagation, the influence of the randomicity of the prebreakdown duration is taken into consideration. By increasing the conductivity of water, the randomicity reduces effectively. Experimental results show that, for a spark discharge, the process can be separated into three stages： the generation and propagation of a streamer, the generation and expansion of the discharge channel, and the development and annihilation of the plasma. The streamers do not directly move to the opposite electrode, but form a bush-like figure. With the increase of the number of branches, the velocity of streamer propagation slows down. The trajectory of the initial channel between electrodes is not straight. However, with the channel expanding, its shape transforms into a straight column. For a corona discharge, there are two stages： the generation and propagation of a streamer, and the stagnation and annihilation of the streamer. The initial streamer in a corona discharge is generated later than in a spark discharge. The forms of streamers for both kinds of discharge are similar; however, streamers generated by a corona discharge propagate with a slower velocity and the number of branches is less compared with a spark discharge. When the energy injection stops, the luminescence of plasma inside the discharge channel （spark discharge） or streamers （corona discharge） becomes weaker and weaker, and finally disappears.

Repeated Sparks Produce Free Energy: A Rigorous Mathematical Explanation of the Authors, by the Hidden Ampere Force Law

This is the twinned paper of the increased mass paper, “Gravity is acting at superluminous infinite speed, thus collapsing special theory of relativity of Einstein, an explanation for almost everything (according to the authors’ rigorous mathematical proof and overwhelming experimental evidence)"?[1]. In the present paper, we describe the repeated spark or repeat current in terms of the hidden, Non-Relativistic Longitudinal Cardinal Ampere Force which is not found yet infallible for every case. The impressive finding is the fact that these repeated sparks as we shall see, have no back-emf, but forward-emf and forward-Ampere-motive-force. The only consumption is the resistive load with the very high currents needed. This means its total energy consumption is less than the released energy-output work plus heat, plus the energy recharging the batteries. All these are in excess of the consumed input energy. This means the Repeated sparking is “over-unity” efficiency and that, using super-conductive materials, we shall have constructed the first “perpetual mobile” plus producing extra electrical energy. Several previous journals that we did not mention here their names did not dare to publish our disliked papers of very important and revolutionary results but were not able to counter argue our arguments. They preferred the silence and have not answered us at all for many years now. Even some of them were involved previously with the repeated sparking which is the same topic of our present paper.

Learning Chinese Sparks Passion for Life

“One World,One Family”is a theme that conjures up images of harmony and the desire to bridge cultural divides,so it was a fitting backdrop for the 18th“Chinese Bridge”—Chinese Proficiency Competition for Foreign College Students held in Changsha,Hunan Province this year.

Modulating sparks in a pulse train for repetitive and energy efficient plasma generation

Spark is a widely studied plasma source for active species production;however,it ex-periences unstable transitions(e.g.to a thermal arc)at high frequencies or long pulse durations.In this study,the sparks generated in a pulse train were studied and modulated based on a physics‐corrected deep learning method.Our results show that a highly repeatable and stable spark plasma source can be achieved by automatically adjusting the voltage amplitude according to the discharge frequency in a high‐frequency pulse train within the time scale of the fluid response.The influences of the electron number density increasing mode and modulated driven voltage profiles on the energy efficiencies were also studied.

医学影像学教学模式的现状与实践

医学影像学是临床医学中一门实践性极强辅助学科,同时也是发展迅猛,日新月异的学科,在现代医疗中扮演着越来越重要的角色。文章通过对影像医学教学的基本要求、教学目标和目的进行说明,采用影像存储和传输系统(picture archiving and communication system,PACS)结合以问题为基础的教学法(problem-based learning,PBL)及以案例为基础的教学法(case-based learning,CBL)教学法进行实践。针对非影像专业学员学习时间短、学习任务重,学员影像基础和需求迥异,部分非影像专业学员自身对医学影像学的重视不足以及放射基地重视不足等问题。文章阐述了PACS与PBL、CBL教学法、钉钉、微信及SPARK学习平台等学习沟通平台相结合,并增加专职人员对基地管理、带教老师一对一指导,以及不断加强科室建设、安排名师上课来提高科室影响力等多种混合教学模式的解决对策,来唤起学员学习影像相关知识的内驱力,达到提高学员轮转学习效果的目的。

基于互信息和融合加权的并行深度森林算法

针对大数据环境下并行深度森林算法中存在不相关及冗余特征过多、多粒度扫描不平衡、分类性能不足以及并行化效率低等问题,提出了基于互信息和融合加权的并行深度森林算法(parallel deep forest algorithm based on mutual information and mixed weighting,PDF-MIMW)。首先,在特征降维阶段提出了基于互信息的特征提取策略(feature extraction strategy based on mutual information,FE-MI),结合特征重要性、交互性和冗余性度量过滤原始特征,剔除过多的不相关和冗余特征;接着,在多粒度扫描阶段提出了基于填充的改进多粒度扫描策略(improved multi-granularity scanning strategy based on padding,IMGS-P),对精简后的特征进行填充并对窗口扫描后的子序列进行随机采样,保证多粒度扫描的平衡;其次,在级联森林构建阶段提出了并行子森林构建策略(sub-forest construction strategy based on mixed weighting,SFC-MW),结合Spark框架并行构建加权子森林,提升模型的分类性能;最后,在类向量合并阶段提出基于混合粒子群算法的负载均衡策略(load balancing strategy based on hybrid particle swarm optimization algorithm,LB-HPSO),优化Spark框架中任务节点的负载分配,降低类向量合并时的等待时长,提高模型的并行化效率。实验表明,PDF-MIMW算法的分类效果更佳,同时在大数据环境下的训练效率更高。

基于Spark分布式框架的海量星表数据时序重构方法研究

时序重构是时域天文学中的一个重要数据处理步骤,也是拟合光变曲线、开展时域分析研究的基础。Hadoop、Spark这类MapReduce分布式模型在执行过程中分布式集群节点间的任务比较独立,需要跨节点的数据传输量较少。提出了非阻塞异步执行流程,每个分布式进程完全针对独立天区的数据进行连续处理,而分块边缘的新增天体导致的其他节点的新增证认任务延时批量追加,并且会根据各进程间的进度不同确定追加方式,保证证认计算没有遗漏,从而在提高并发效率的同时保证算法的精度。此外,对两表间的不同Join策略从理论和实验两个角度进行了研究并提出了免Join策略。最后通过基于Spark分布式框架的高效时序重构系统的设计完成了以上研究的验证。实验表明,与以往研究结果相比,该时序重构算法效率提升明显,为时域天文学中的天文时序数据分析的开展打下了良好的基础。

基于Spark Streaming的海量GPS数据实时地图匹配算法

浮动车GPS数据作为交通信息处理的基础,随着被监控车辆数量的高速增长,产生了海量GPS数据,对地图匹配提出了挑战。为了解决传统匹配方法难以满足匹配效率和精度的不足,提出一种针对海量GPS数据的实时并行地图匹配算法,能够同时保证较高匹配精度和运算效率。为构建一种面向实时数据流的高效、准确实时地图匹配算法,首先通过引入速度、方向综合权重因子对依赖历史轨迹的离线地图匹配算法进行重构,进而引入Spark Streaming分布式计算框架,实现地图匹配算法的实时、并行运算,大幅提升实时地图匹配效率。实验结果表明,该算法在复杂路段的匹配准确率较常规拓扑匹配算法提高10%以上,整体匹配准确率达到95%以上;在匹配效率方面,较同等数量的单机服务器效率可提高4倍左右。实验结果表明,该算法在由11台机器组成的计算集群上实现8000万个GPS数据点的实时地图匹配,证明了该算法可以完成城市地区的实时车辆匹配。

基于Spark和NRSCA策略的并行深度森林算法

针对并行深度森林在大数据环境下存在冗余及无关特征过多、两端特征利用率过低、模型收敛速度慢以及级联森林并行效率低等问题,提出了基于Spark和NRSCA策略的并行深度森林算法——PDF-SNRSCA。首先,该算法提出了基于邻域粗糙集和Fisher score的特征选择策略(FS-NRS),通过衡量特征的相关性和冗余度,对特征进行过滤,有效减少了冗余及无关特征的数量;其次,提出了一种随机选择和等距提取的扫描策略(S-RSEE),保证了所有特征能够同概率被利用,解决了多粒度扫描两端特征利用率低的问题;最后,结合Spark框架,实现级联森林并行化训练,提出了基于重要性指数的特征筛选机制(FFM-II),筛选出非关键性特征,平衡增强类向量与原始类向量维度,从而加快模型收敛速度,同时设计了基于SCA的任务调度机制(TSM-SCA),将任务重新分配,保证集群负载均衡,解决了级联森林并行效率低的问题。实验表明,PDF-SNRSCA算法能有效提高深度森林的分类效果,且对深度森林并行化训练的效率也有大幅提升。

大数据环境下的分布式机器学习框架比较研究

在大数据环境下,处理庞大的数据集与实现复杂的机器学习算法愈发关键。为解决这一挑战,分布式机器学习框架应运而生。通过分布式计算资源的协同工作,可以提高机器学习模型的训练效率和性能。

Spark综合实验平台的使用和实验教学实践

Spark实验课是Spark理论课程的延伸和拓展,为了提高学生学习Spark实验课程的兴趣,文章对Spark综合实验平台的实践教学进行了探索,实践教学的结果表明,采用Spark综合实验平台进行教学,提高了学生的学习兴趣和求知欲,理论课和实验课的成绩都得到了提高。

失物招领处:寻找1‰秒的Soul Sparks

我是一名外科医生,今年30岁。30岁生日那天我在微博上写了一篇文章,题目是《你还记得那个青年医生吗?》。孔老夫子所说的"三十而立",我们究竟应该如何理解呢?是发一篇牛气闪闪的SCI吗,还是申请一项超过百万级别的科研基金呢,还是独立操作达芬奇机器人做一台手术呢,或者,我们换一栋大房子,买一辆心仪已久的跑车,还是带着心爱的姑娘去周游世界?像我们这样,30岁上下,夹在现实与理想之间,父母和孩子之间。

基于大数据随机样本划分的分布式观测点分类器

观测点分类器(OPC)是一种试图通过将多维样本空间线性不可分问题转换成一维距离空间线性可分问题的有监督学习模型,对高维数据的分类问题尤为有效。针对OPC在处理大数据分类问题时表现的较高训练复杂度,在Spark框架下设计一款基于大数据的随机样本划分(RSP)的分布式OPC(DOPC)。首先,在分布式计算环境下生成大数据的RSP数据块,并将它转换为弹性分布式数据集(RDD);其次,在RSP数据块上协同式地训练一组OPC,由于每个RSP数据块上的OPC独立训练,因此有高效的Spark可实现性;最后,在Spark框架下将在RSP数据块上协同训练的OPC集成为DOPC,对新样本进行类标签预测。在8个大数据集上,对Spark集群环境下实现的DOPC的可行性、合理性和有效性进行实验验证,实验结果显示,DOPC能够以更低的计算消耗获得比单机OPC更高的测试精度,同时相较于Spark框架下实现的基于RSP模型的神经网络(NN)、决策树(DT)、朴素贝叶斯(NB)和K最近邻(KNN),DOPC分类器具有更强的泛化性能。测试结果表明,DOPC是一种高效低耗的处理大数据分类问题的有监督学习算法。

基于Spark平台的并行化谱聚类算法的在线学习资源推荐

为了提高在线学习资源推荐的准确度,采用谱聚类用于学习资源的归类,将类别相似度高的资源推荐给用户,提出Spark平台的并行化谱聚类算法,提高资源推荐效率;首先提取在线学习资源及用户特征并初始化,建立谱聚类模型,在Spark平台上分别求解无向图的顶点相似度及归一化拉普拉斯系数;然后采用归一化分割划分子集,通过归一化割集优化方式求解类别特征,并对类别特征按行输出特征点;最后采用k均值算法对特征点进行聚类,获得聚类结果。结果表明,采用谱聚类算法并借助于Spark平台的计算优势,所提推荐方法比常用的在线学习资源推荐算法的准确率和覆盖率更高,在海量学习资源的实时推荐方面具有较高适应度。

面向结构化篇级科技文献数据治理的高性能分布式计算框架研究

[研究目的]为解决MapReduce、Spark等主流分布式计算框架存在的研发周期长、技术门槛高等问题,提出了一种高灵活、低门槛的高性能计算框架ArticleCF。[研究方法]ArticleCF框架吸收了主流分布式技术的优点,同时深度结合科技文献数据治理的特性,设计了Master/Slave的软件架构,在功能上针对科技文献数据特点进行多个维度的设计,重点设计了分布式任务分发策略、并行计算策略以及故障转移机制。[研究结论]通过21个指标将ArticleCF与MapReduce、Spark、Storm进行对比实验,有效验证所提方法的可行性、有效性,ArticleCF能够满足海量结构化科技文献数据的多样化处理需求。