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%。     

深部松软煤层钻孔孔周煤体变形产渣特征研究

针对煤矿井下高瓦斯软煤顺层长钻孔排渣困难、成孔率低、施工困难等问题,通过数值模拟实验研究了井下深部软煤体变形破坏特征,分析了顺层长钻孔孔周松软煤体变形特征及应力变化,以揭示顺层长钻孔孔周松软煤体变形产渣规律。研究表明:深部高瓦斯软煤顺层钻孔孔周煤体的应力平衡临界条件破坏后将发生大体积突然垮落;钻孔水平最大变形位移为1.22mm,垂直方向最大变形位移为10.7mm;径向孔周煤体垂向变形呈现逐渐减小趋势,且垂向变形明显大于钻孔水平变形。在水平方向上,钻孔孔周煤体应力分布呈现先增大再逐渐减小的变化规律,径向距离对水平应力分布的影响逐渐减小;随着径向距离的增加,钻孔孔周煤体应力分布逐渐降低,钻孔孔壁处煤体的应力出现最大值,且垂直方向处应力值最大。     

NNI-1型C_5/C_6烷烃异构化催化剂长周期运行分析及建议

中国石化海南炼油化工有限公司0.2 Mt/a C5/C6烷烃异构化装置以连续重整装置的拔头油为原料,使用NNI-1催化剂,采用一次通过流程,不设脱异戊烷塔和稳定塔,经设在连续重整装置内的脱丁烷塔稳定处理后作为汽油调合组分。该装置于2006年9月开工投产,截至2015年3月已连续运行3个周期。长周期运行分析结果表明:前两个周期中NNI-1催化剂具有较高的异构化活性及选择性,C5异构化率为60%左右,C6异构化率为80%左右,C6选择性为15%左右,产品辛烷值基本达到技术指标要求(RON≥78);而在第三周期运行中,催化剂积炭增加等原因导致其异构化活性及选择性降低,异构化产品辛烷值提升能力呈现逐步衰减的趋势,提高反应苛刻度已不能弥补催化剂活性下降造成的产品辛烷值降低。为保证装置长周期运行,建议择机停工对催化剂进行再生,或是直接换用与装置原料性质匹配的异构化催化剂。     

In vivo Kinetic Biodistribution of Nano‐Sized Outer Membrane Vesicles Derived from Bacteria

Evaluation of kinetic distribution and behaviors of nanoparticles in vivo provides crucial clues into their roles in living organisms. Extracellular vesicles are evolutionary conserved nanoparticles, known to play important biological functions in intercellular, inter‐species, and inter‐kingdom communication. In this study, the first kinetic analysis of the biodistribution of outer membrane vesicles (OMVs)—bacterial extracellular vesicles—with immune‐modulatory functions is performed. OMVs, injected intraperitoneally, spread to the whole mouse body and accumulate in the liver, lung, spleen, and kidney within 3 h of administration. As an early systemic inflammation response, increased levels of TNF‐α and IL‐6 are observed in serum and bronchoalveolar lavage fluid. In addition, the number of leukocytes and platelets in the blood is decreased. OMVs and cytokine concentrations, as well as body temperature are gradually decreased 6 h after OMV injection, in concomitance with the formation of eye exudates, and of an increase in ICAM‐1 levels in the lung. Following OMV elimination, most of the inflammatory signs are reverted, 12 h post‐injection. However, leukocytes in bronchoalveolar lavage fluid are increased as a late reaction. Taken together, these results suggest that OMVs are effective mediators of long distance communication in vivo.     

Multiple Empirical Kernel Learning with Majority Projection for imbalanced problems

Traditional Multiple Empirical Kernel Learning (MEKL) expands the expressions of the sample and brings better classification ability by using different empirical kernels to map the original data space into multiple kernel spaces. To make MEKL suit for the imbalanced problems, this paper introduces a weight matrix and a regularization term into MEKL. The weight matrix assigns high misclassification cost to the minority samples to balanced misclassification cost between minority and majority class. The regularization term named Majority Projection (MP) is used to make the classification hyperplane fit the distribution shape of majority samples and enlarge the between-class distance of minority and majority class. The contributions of this work are: (i) assigning high cost to minority samples to deal with imbalanced problems, (ii) introducing a new regularization term to concern the property of data distribution, (iii) and modifying the original PAC-Bayes bound to test the error upper bound of MEKL-MP. Through analyzing the experimental results, the proposed MEKL-MP is well suited to the imbalanced problems and has lower generalization risk in accordance with the value of PAC-Bayes bound.     

Functional Dysregulations in CA1 Hippocampal Networks of a 3-Hit Mouse Model of Schizophrenia

For a better translation from treatment designs of schizophrenia to clinical efficiency, there is a crucial need to refine preclinical animal models. In order to consider the multifactorial nature of the disorder, a new mouse model associating three factors (genetic susceptibility—partial deletion of the MAP6 gene, early-life stress—maternal separation, and pharmacological treatment—chronic Δ-9-tetrahydrocannabinol during adolescence) has recently been described. While this model depicts a schizophrenia-like phenotype, the neurobiological correlates remain unknown. Synaptic transmission and functional plasticity of the CA1 hippocampal region of male and female 3-hit mice were therefore investigated using electrophysiological recordings on the hippocampus slice. While basal excitatory transmission remained unaffected, NMDA receptor (NMDAr)-mediated long-term potentiation (LTP) triggered by theta-burst (TBS) but not by high-frequency (HFS) stimulation was impaired in 3-hit mice. Isolated NMDAr activation was not affected or even increased in female 3-hit mice, revealing a sexual dimorphism. Considering that the regulation of LTP is more prone to inhibitory tone if triggered by TBS than by HFS, the weaker potentiation in 3-hit mice suggests a deficiency of intrinsic GABA regulatory mechanisms. Indeed, NMDAr activation was increased by GABA_A receptor blockade in wild-type but not in 3-hit mice. This electrophysiological study highlights dysregulations of functional properties and plasticity in hippocampal networks of 3-hit mice, one of the mechanisms suspected to contribute to the pathophysiology of schizophrenia. It also shows differences between males and females, supporting the sexual dimorphism observed in the disorder. Combined with the previously reported study, the present data reinforce the face validity of the 3-hit model that will help to consider new therapeutic strategies for psychosis.     

基于粒子群优化和支持向量机的花粉浓度预测模型

为了提高花粉浓度预报的准确率，解决现有花粉浓度预报准确率不高的问题，提出了一种基于粒子群优化（PSO）算法和支持向量机（SVM）的花粉浓度预报模型。首先，综合考虑气温、气温日较差、相对湿度、降水量、风力、日照时数等多种气象要素，选择与花粉浓度相关性较强的气象要素构成特征向量；其次，利用特征向量与花粉浓度数据建立SVM预测模型，并使用PSO算法找出最优参数；然后利用最优参数优化花粉浓度预测模型；最后，使用优化后的模型对花粉未来24 h浓度进行预测，并与未优化的SVM、多元线性回归法（MLR）、反向神经网络（BPNN）作对比。此外使用优化后的模型对某市南郊观象台和密云两个站点进行逐日花粉浓度预测。实验结果表明，相比其他预报方法，所提方法能有效提高花粉浓度未来24 h预测精度，并具有较高的泛化能力。     

Backtesting portfolio value-at-risk with estimated portfolio weights

This article theoretically and empirically analyzes backtesting portfolio value-at-risk (VaR) with estimation risk in an intrinsically multi-variate framework. It particularly takes into account the estimation of portfolio weights in forecasting portfolio VaR and its impact on backtesting. It shows that the estimation risk from estimating portfolio weights and that from estimating the multi-variate dynamic model make the existing methods in a univariate framework inapplicable. It proposes a general theory to quantify estimation risk applicable to the present problem and suggests practitioners a simple but effective way to implement valid inference to overcome the effect of estimation risk in backtesting portfolio VaR. In particular, we apply our theory to the efficient mean-variance-skewness portfolio for a multi-variate generalized autoregressive conditional heteroscedasticity model with multi-variate general hyperbolic distributed innovations. Some Monte Carlo simulations and an empirical application demonstrate the merits of our method.     

Learning keyword phrases from query logs of USPTO patent examiners for automatic query scope limitation in patent searching

Professional search in patent repositories poses several unique challenges. One key requirement is to search the entire affected space of concepts, following well-defined procedures to ensure traceability of results obtained. Several techniques have been introduced to enhance query generation, preferably via automated query term expansion, to improve retrieval effectiveness. Currently, these approaches are mostly limited to computing additional query terms from patent documents based on statistical measures. For conceptual search to solve the limitation of traditional keyword search standard dictionaries are used to provide synonyms and keyword phrases for query refinement. Studies show that these are insufficient in such highly specialized domains. In this paper, we present an approach to extract keyword phrases from query logs created during the validation procedure of the patent applications. This creates valuable domain-specific lexical databases for several specific patent classes that can be used to both expand as well as limit the scope of a patent search. This provides a more powerful means to guide a professional searcher through the search process. We evaluate the lexical databases based on real query sessions of patent examiners.