简介:PRAreferstotheprocessofself-conscienceandself-developmentofacommunitypushedbyexternalworkersusingPRAtool.Itlaysstressontheparticipationoflocalworkers,thecoordinationandassistancebyexternalworkerstolocalpeopletocarryoutinvestigationandanalysis,andthesharingofresults.Sincethismethodisproducedintheform
简介:当察觉到环境时,一个类人动物机器人被察觉到的信息总是充满,;计算通常需要重要时间;处理察觉到的信息。在这篇论文,一条选择基于注意的上下文的感觉途径被建议让类人动物机器人与高效率察觉到环境。首先,注意窗户(AW)的涵义被扩大做一个更多的将军;AW的抽象定义,;它的四种操作;声明转变也被讨论。第二,注意控制政策被描述,它集成指导目的的感性的对象选择;错误答案抑制,;能处理突现的问题。错误答案抑制被用来过滤无关的信息。最后,注意政策被看作机器人的感性的模式,罐头控制它;调整感觉效率。试验性的结果证明介绍途径能显著地支持感性的效率,;感性的费用能有效地通过采用不同注意政策被控制。
简介:ZouYan,female,researcher,majoringinevidence-basedmedicine,graduatedfromWestChinaSchoolofMedicine,SichuanUniversityandgainedadegreeofMedicalDoctor.Shehadbeenengagedinclinicalworkofobstetricsandgynecologyfortenyears.Currently,sheworksasHeadofWomenClinicalLaboratoryintheScientificResearchInstituteofNationalHealthandFamilyPlanningCommission,andamasterinstructorofDepartmentofObstetricsandGynecologyinPekingUnionMedicalCollege.
简介:Inthispaper,anovelframework,namedasglobal-localfeatureattentionnetworkwithrerankingstrategy(GLAN-RS),ispresentedforimagecaptioningtask.Ratherthanonlyadoptingunitaryvisualinformationintheclassicalmodels,GLAN-RSexplorestheattentionmechanismtocapturelocalconvolutionalsalientimagemaps.Furthermore,weadoptrerankingstrategytoadjustthepriorityofthecandidatecaptionsandselectthebestone.TheproposedmodelisverifiedusingtheMicrosoftCommonObjectsinContext(MSCOCO)benchmarkdatasetacrosssevenstandardevaluationmetrics.ExperimentalresultsshowthatGLAN-RSsignificantlyoutperformsthestate-of-the-artapproaches,suchasmultimodalrecurrentneuralnetwork(MRNN)andGoogleNIC,whichgetsanimprovementof20%intermsofBLEU4scoreand13pointsintermsofCIDERscore.
简介:AbstractPurpose:Given the increased risk of accidents in patients with attention deficit and hyperactivity disorder (ADHD) or maternal anxiety/depression, we aimed to investigate the frequency of the two diseases in children with penetrating eye injury (PEI).Methods:Altogether 79 children, 39 with PEIs and 40 healthy individuals (control group), aged 5-15 years, underwent a complete ophthalmologic examination. Afterwards, schedule for affective disorders and schizophrenia for school-aged children was conducted to assess the psychiatric diagnosis of all children. Turgay diagnostic and statistical manual of mental disorders (DSM-IV)-based child and adolescent behavior disorders screening and rating scale (T-DSM-IV-S) was filled by parents to evaluate the severity of ADHD symptoms. The depression and anxiety levels of mothers of each group were evaluated by two self-report measures: the Beck depression scale and the state-trait anxiety inventory (STAI), respectively. Data were analyzed by IBM SPSS version 22.0. The Chi-square and Fisher's exact test were used to determine whether there is a significant difference between qualitative variables while independent sample t and Mann-Whitney U tests to compare quantitative variables.Results:The only diagnostic difference was a significantly higher frequency of ADHD among patients with PEIs (48.7% in PEI vs. 17.5% in control group, χ2 = 7.359, p = 0.007). The total scores of the T-DSM-IVS (attention subscale U = 418.000, p = 0.006; hyperactivity subscale U = 472.000, p = 0.022) and maternal state-trait anxiety inventory (maternal STAI-state U = 243.000, p = 0.003; maternal STAI-trait U = 298.000, p = 0.021) were significantly higher in the PEI group than in control group. In logistic regression, children with PEI had a tendency to have a 3.5-fold increased risk for ADHD (OR = 3.538, CI = 0.960-13.039, p = 0.058).Conclusion:ADHD was detected almost 1 in 2 children with PEIs. Besides, the maternal anxiety level was significantly higher in the PEI group than in the control group. This association should be further explored via a future prospective longitudinal study. Since a proper treatment of ADHD in children and anxiety treatment in mothers may prevent vision loss following PEIs in children.
简介:
简介:摘要:糖尿病视网膜病变 (DR) 是一种严重的眼部异常,其严重情况下会导致视网膜脱落甚至失明。眼底的渗出液是由于高血糖毒性作用,导致血屏障破坏,血管内的脂质等漏出而造成的。是视网膜病变的并发症之一。由于患者与专业医生数量悬殊巨大,设计一个可以自动的检测渗出液的医疗助手是十分重要的任务。本文依托于深度学习方法,以U-Net架构为骨架网络,以准确度 (Acc)、灵敏度 (SE)、特异性 (SP)以及AUC值作为模型性能的评估指标,先测试了原始U-Net在该任务上的分割能力,在该任务上达到99.8%的准确度,73.1%的灵敏度,98.0%的特异性以及0.973的AUC值。根据U-Net网络架构的固有问题,将Attention机制与U-Net结构,搭建Attention U-Net。99.8%的准确度,81.5%的灵敏度,99.8%的特异性以及0.985的AUC值。实验结果表明,Attention U-Net有更好的特征提取能力。
简介: