基于多通道的ICU脑血管疾病死亡风险预测模型

死亡风险预测指根据病人临床体征监测数据来预测未来一段时间的死亡风险。对于ICU病患，通过死亡风险预测可以有针对性地对病人做出临床诊断，以及合理安排有限的医疗资源。基于临床使用的MEWS和Glasgow昏迷评分量表，针对ICU病人临床监测的17项生理参数，提出一种基于多通道的ICU脑血管疾病死亡风险预测模型。引入多通道概念应用于BiLSTM模型，用于突出每个生理参数对死亡风险预测的作用。采用Attention机制用于提高模型预测精度。实验数据来自MIMIC [Ⅲ]数据库，从中提取3?080位脑血管疾病患者的16?260条记录用于此次研究，除了六组超参数实验之外，将所提模型与LSTM、Multichannel-BiLSTM、逻辑回归（logistic regression）和支持向量机（support vector machine, SVM）四种模型进行了对比分析，准确率Accuracy、灵敏度Sensitive、特异性Specificity、AUC-ROC和AUC-PRC作为评价指标，实验结果表明，所提模型性能优于其他模型，AUC值达到94.3%。     

基于推理注意力机制的二阶段网络图像修复

现有的图像修复方法在处理大面积缺失或高度纹理化的图像时，通常会产生扭曲的结构或与周围区域不一致的模糊纹理，无法重建合理的图像结构。为此，提出了一种基于推理注意力机制的二阶段网络图像修复方法。首先通过边缘生成网络生成合理的幻觉边缘信息，然后在图像补全网络完成图像的重建工作。为了进一步生成视觉效果更逼真的图像，提高图像修复的精确度，在图像补全网络采用推理注意力机制，有效控制了生成特征的不一致性，从而生成更有效的信息。所提方法在多个数据集上进行了实验验证，结果表明该图像修复方法的结构相似性指数达到了88.9%，峰值信噪比达到了25.56 dB，与现有的图像修复方法相比，该方法具有更高的图像修复精确度，生成的图像更逼真。     

基于阶段注意力机制的电力线提取算法

电力系统维护是电力系统稳定运行的重要保障，应用智能算法的无人机电力巡检则为电力系统维护提供便捷。电力线提取是自主电力巡检以及保障飞行器低空飞行安全的关键技术，结合深度学习理论进行电力线提取是电力巡检的重要突破点。本文将深度学习方法用于电力线提取任务，结合电力线图像特点嵌入改进的图像输入策略和注意力模块，提出一种基于阶段注意力机制的电力线提取模型（SA-Unet）。本文提出的SA-Unet模型编码阶段采用阶段输入融合策略（Stage input fusion strategy， SIFS），充分利用图像的多尺度信息减少空间位置信息丢失。解码阶段通过嵌入阶段注意力模块（Stage attention module，SAM）聚焦电力线特征，从大量信息中快速筛选出高价值信息。实验结果表明，该方法在复杂背景的多场景中具有良好的性能。     

基于注意力机制与文本信息的用户关系抽取

随着社交媒体的发展,用户之间的关系网络对于社交媒体的分析有很大的帮助。因此,该文主要研究用户好友关系检测。以往的关于用户好友关系抽取的研究主要基于社交媒体上的结构化信息,比如其他好友关系,用户的不同属性等。但是,很多时候用户本身并没有大量的好友信息存在,同时也不一定有很多确定的属性。因此,我们希望基于用户发表的文本信息来对用户关系进行预测。不同于以往的潜在好友推荐算法,该文提出了一种基于注意力机制以及长短时记忆网络(long short-term memory,LSTM)的好友关系预测模型,将好友之间的评论分开处理,通过分析用户之间的评论来判断是否具备一定的好友关系。该模型将好友双方信息拼接后的结果作为输入,并将注意力机制应用于LSTM的输出。实验表明,用户之间的评论对于好友关系预测确实有较大的实际意义,该文提出的模型较之于多个基准系统的效果,取得了明显的提升。在不加入任何其它非文本特征的情况下,实验结果的准确率达到了77%。     

Context-Processing Deficits in Schizotypal Personality Disorder.

Research suggests that schizotypal personality disorder (SPD) is a part of the spectrum of schizophrenia-related illnesses. This article hypothesizes that a deficit in the representation and maintenance of context is a core cognitive disturbance in schizophrenia and that SPD individuals should demonstrate context-processing deficits. To test this hypothesis, the authors administered 3 versions of their AX-CPT task, designed to assess context processing, to 35 healthy controls and 26 individuals with DSM-IV SPD. They also administered working memory and selective attention tasks. SPD individuals displayed context representation deficits similar to those found in schizophrenia but did not show the same additional deficits in context maintenance. Context processing was strongly associated with working memory and selective attention performance in the SPD individuals. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The accuracy of self-monitoring and its relationship to self-focused attention in dysphoria and clinical depression.

The accuracy with which dysphoric (Study 1) and clinically depressed (Study 2) individuals make self-regulatory judgments about their own performance in the absence of external feedback and the extent to which this relates to trait self-focused attention (SFA) were examined. Relative to objective criteria, both dysphoric and depressed participants showed a positive judgment bias, overestimating the number of trials they had performed correctly. Relative to control participants, the dysphoric and depressed groups showed a reduction in the extent of this positive bias in that they judged error trials more accurately and correct trials less accurately. Although the dysphoric and depressed groups both reported elevated trait SFA, this did not correlate significantly with accuracy of self-judgment on the performance-monitoring task. Implications for self-regulation models of depression are discussed. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Serial deployment of attention during visual search.

This study examined whether objects are attended in serial or in parallel during a demanding visual search task. A component of the event-related potential waveform, the N2pc wave, was used as a continuous measure of the allocation of attention to possible targets in the search arrays. Experiment 1 demonstrated that the relative allocation of attention shifts rapidly, favoring one item and then another. In Experiment 2, a paradigm was used that made it possible to track the absolute allocation of attention to individual items. This experiment showed that attention was allocated to one object for 100-150 ms before attention began to be allocated to the next object. These findings support models of attention that posit serial processing in demanding visual search tasks. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Cutting stress off at the pass: Reducing vigilance and responsiveness to social threat by manipulating attention.

Personality processes relating to social perception have been shown to play a significant role in the experience of stress. In 5 studies, the authors demonstrate that early stage attentional processes influence the perception of social threat and modify the human stress response. The authors first show that cortisol release in response to a stressful situation correlates with selective attention toward social threat. Second, the authors show in 2 laboratory studies that this attentional pattern, most evident among individuals with low self-esteem, can be modified with a repetitive training task. Next, in a field study, students trained to modify their attentional pattern to reduce vigilance for social threat showed lower self-reported stress related to their final exam. In a final field study with telemarketers, the attentional training task led to increased self-esteem, decreased cortisol and perceived stress responses, higher confidence, and greater work performance. Taken together, these results demonstrate the impact of antecedent-focused strategies on the late-stage consequences of social stress. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Symbolic control of visual attention: The role of working memory and attentional control settings.

This study examined how 1 symbol is selected to control the allocation of attention when several symbols appear in the visual field. In Experiments 1-3, the critical target feature was color, and it was found that uninformative central arrows that matched the color of the target were selected and produced unintentional shifts of attention (i.e., involuntary, initiated slowly, producing long-lasting facilitatory effects). Experiment 4 tested whether such selection is the result of an attentional filter or of a competition bias due to a match of incoming information against integrated object representations stored in working memory. Here, the critical feature was shape and color was irrelevant, but matching color arrows were still selected. Thus, features of objects in working memory will bias the selection of symbols in the visual field, and such selected symbols are capable of producing unintentional shifts of attention. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Cognitive Operations in the Generation Effect on a Recall Test: Role of Aging and Divided Attention.

Generation effect (generated words are better memorized than read words) of anagrams, rhymes, and associates of target words was examined in young, elderly, and very old subjects. Experiments 1 and 2 showed that only young subjects benefit from the generation effect in a free-recall test when the rule is of a phonological nature. Experiments 3, 4, and 5 showed that the generation effect of rhymes was due to a resources-dependent self-initiated process. Experiments 4 and 5 showed that in a divided-attention situation, generation effect of rhymes is not significant in young subjects, but that the generation effect of semantic associates remains significant for both groups (Experiment 5). The results are discussed within the environmental support framework and the transfer-appropriate processing framework. (PsycINFO Database Record (c) 2010 APA, all rights reserved)