Depression recognition using functional connectivity based on dynamic causal model

Dynamic casual modeling of functional magnetic resonance imaging（fMRI） signals is employed to explore critical emotional neurocircuitry under sad stimuli. The intrinsic model of emotional loops is built on the basis of Papez＇s circuit and related prior knowledge, and then three modulatory connection models are established. In these models, stimuli are placed at different points, which represents they affect the neural activities between brain regions, and these activities are modulated in different ways. Then, the optimal model is selected by Bayesian model comparison. From group analysis, patients＇ intrinsic and modulatory connections from the anterior cingulate cortex （ACC） to the right inferior frontal gyrus （rlFG） are significantly higher than those of the control group. Then the functional connection parameters of the model are selected as classifier features. The classification accuracy rate from the support vector machine（SVM） classifier is 80.73%, which, to some extent, validates the effectiveness of the regional connectivity parameters for depression recognition and provides a new approach for the clinical diagnosis of depression.

Observed communication between oncologists and patients:A causal model of communication competence

AIM： To investigate and test a causal model derivedfrom previous meta-analytic data of health provider be-haviors and patient satisfaction.METHODS： A literature search was conducted forrelevant manuscripts that met the following criteria：Reported an analysis of provider-patient interaction inthe context of an oncology interview; the study hadto measure at least two of the variables of interest tothe model （provider activity, provider patient-centeredcommunication, provider facilitative communication,patient activity, patient involvement, and patient satis-faction or reduced anxiety）; and the information had tobe reported in a manner that permitted the calculationof a zero-order correlation between at least two of thevariables under consideration. Data were transformedinto correlation coefficients and compiled to producethe correlation matrix used for data analysis. The test of the causal model is a comparison of the expected correlation matrix generated using an Ordinary Least Squares method of estimation. The expected matrix iscompared to the actual matrix of zero order correlation coeffcients. A model is considered a possible ft if the level of deviation is less than expected due to random sampling error as measured by a chi-square statistic. The signifcance of the path coeffcients was tested us-ing a z test. Lastly, the Sobel test provides a test of the level of mediation provided by a variable and provides an estimate of the level of mediation for each connec-tion. Such a test is warranted in models with multiple paths.RESULTS： A test of the original model indicated a lack of ft with the summary data. The largest discrepancy in the model was between the patient satisfaction and the provider patient-centered utterances. The observed correlation was far larger than expected given a medi-ated relationship. The test of a modifed model was un-dertaken to determine possible ft. The corrected model provides a fit to within tolerance as evaluated by the test statistic, χ2 （8, average n = 342） = 10.22. Each of the path coefficients for the model reveals that each one can be considered signifcant, P 〈 0.05. The Sobel test examining the impact of the mediating variables demonstrated that patient involvement is a signifcantmediator in the model, Sobel statistic = 3.56, P 〈 0.05. Patient active was also demonstrated to be a signifcant mediator in the model, Sobel statistic = 4.21, P 〈 0.05. The statistics indicate that patient behavior mediates the relationship between provider behavior and patient satisfaction with the interaction.CONCLUSION： The results demonstrate empirical support for the importance of patient-centered care and satisfy the need for empirical casual support of provider-patient behaviors on health outcomes.

ICA Based Identification of Time-Varying Linear Causal Model

Recently, several approaches have been proposed to discover the causality of the time-independent or fixed causal model. However, in many realistic applications, especially in economics and neuroscience, causality among variables might be time-varying. A time-varying linear causal model with non-Gaussian noise is considered and the estimation of the causal model from observational data is focused. Firstly, an independent component analysis(ICA) based two stage method is proposed to estimate the time-varying causal coefficients. It shows that, under appropriate assumptions, the time varying coefficients in the proposed model can be estimated by the proposed approach, and results of experiment on artificial data show the effectiveness of the proposed approach. And then, the granger causality test is used to ascertain the causal direction among the variables. Finally, the new approach is applied to the real stock data to identify the causality among three stock indices and the result is consistent with common sense.

Application of causal model to maternal smoking cessation intervention in pregnancy

The adverse effects of maternal smoking during pregnancy on both the offspring and women are well known. The main objective of this research article is to provide health professional causal modelling approach to make a more comprehensive assessment of major determinants of smoking behaviour during and after pregnancy and consequently the outcomes of pregnant women smoking which are adversely affecting both the offspring and pregnant women. The causal model based on theory and evidence was modified and applied to material smoking cessation intervention to control the adverse effects of smoking on offspring obesity and neurodevelopment. In this approach a generic model links behavioural determinants, causally through behaviour, to physiological and biochemical variables, and health outcomes. It is tailored to context, target population, behaviours and health outcomes. The model provides a rational guide to appropriate measures, intervention points and intervention techniques, and can be tested quantitatively. The causal modelling approach showed promising results which can be used to help maternal smoking women to understand the risk of smoking and help them to quit smoking. The regression analysis of maternal smoking women BMI (n = 1000) on offspring BMI was statistically significant, p 0.05). This supported the hypothesis that maternal smoking women BMI during pregnancy is an important determinant of offspring obesity and consequently the risk factors of cardiovascular development. The causal modelling approach is unique as it provides an incentive to health professional to use these models to target any important and modifiable determinants of the maternal smoking behaviour and decrease the risk of adverse pregnancy outcomes for the offspring and the mother.

Using granger-geweke causality model to evaluate the effective connectivity of primary motor cortex, supplementary motor area and cerebellum

Currently, Granger-Geweke causality models have been widely applied to investigate the dynamic direction relationships among brain regions. In a previous study, we have found that the right hand finger-tapping task can produce relatively reliable brain response. As an extension of our previous study, we developed an algorithm based on the classical Granger- Geweke causality model to further investigate the effective connectivity of three brain regions (left primary motor cortex (M1), supplementary motor area (SMA) and right cerebellum) that showed the most robust brain activations. Our computational results not only confirm the strong linear feedback among SMA, M1 and right cerebellum, but also demonstrate that M1 is the hub of these three regions indicated by the anatomy research. Moreover, the model predicts the high intermediate node density existing in the area between SMA and M1, which will stimulate the imaging experimentalists to carry out new experiments to validate this postulation.

基于Modelica语言的电液伺服阀非因果建模仿真

针对电液伺服阀这一典型机械、电子、磁场、液压与控制耦合的复杂系统,考虑其多领域、多层次化的特点,应用多领域建模仿真语言Modelica对电液伺服阀进行建模仿真。不同于传统的电液伺服阀控制框图建模分析方法,Modelica语言的非因果性与面向对象性,使得该方法具有强可读性,强可用性,便于修改等特点,克服了传统方法不能从底层元件反映伺服阀特性、考虑因果性、不易修改与重用的缺点。仿真结果显示,该方法可以反映电液伺服阀的动态特性,同时可以方便地通过参数修正对电液伺服阀进行优化设计,为电液伺服工程技术人员提供了电液伺服阀设计的更为高效便捷的手段。

基于Modelica的MEMS系统级多领域建模与仿真

分析了现有的MEMS系统级建模与仿真方法,讨论了运用Modelica语言进行面向对象的非因果关系建模方法,建立了基于Modelica的电容式微型静电致动器系统级模型,仿真结果证明了Modelica用于MEMS系统级多领域仿真的可行性。

Robust Parameter Identification Method of Adhesion Model for Heavy Haul Trains

A robust parameter identification method based on Kiencke model was proposed to solve the problem of the parameter identification accuracy being affected by the rail environment change and noise interference for heavy-duty trains. Firstly, a Kiencke stick-creep identification model was constructed, and the parameter identification task was transformed into a quadratic programming problem. Secondly, an iterative algorithm was constructed to solve the problem, into which a time-varying forgetting factor was added to track the change of the rail environment, and to solve the uncertainty problem of the wheel-rail environment. The Granger causality test was adopted to detect the interference, and then the weights of the current data were redistributed to solve the problem of noise interference in parameter identification. Finally, simulations were carried out and the results showed that the proposed method could track the change of the track environment in time, reduce the noise interference in the identification process, and effectively identify the adhesion performance parameters.

基于Modelica/Dymola的微谐振器建模与仿真

针对MEMS的多领域耦合和系统级的快速建模与仿真要求,提出研究关于微型梳状静电谐振器建模与仿真方法。为提高系统的稳定性,减少误差,采用了基于Modelica/Dymola的非因果关系建模方法及流程,以梳状谐振器机电耦合模型的自然形式方程为基础,以Modelica语言建立非因果关系的仿真模型,借助于Dymola平台对振子的位移/速度曲线进行仿真。仿真结果与理论验证相符,表明Modelica具有建模过程简单、建模速度快和仿真精度高等优点,适合于MEMS多领域建模与仿真研究。

Causal association rule mining methods based on fuzzy state description

Aiming at the research that using more new knowledge to develope knowledge system with dynamic accordance, and under the background of using Fuzzy language field and Fuzzy language values structure as description framework, the generalized cell Automation that can synthetically process fuzzy indeterminacy and random indeterminacy and generalized inductive logic causal model is brought forward. On this basis, a kind of the new method that can discover causal association rules is provded. According to the causal information of standard sample space and commonly sample space, through constructing its state （abnormality） relation matrix, causal association rules can be gained by using inductive reasoning mechanism. The estimate of this algorithm complexity is given,and its validiw is proved through case.

Causal inference using regression-based statistical control: Confusion in Econometrics

Regression is a widely used econometric tool in research. In observational studies, based on a number of assumptions, regression-based statistical control methods attempt to analyze the causation between treatment and outcome by adding control variables. However, this approach may not produce reliable estimates of causal effects. In addition to the shortcomings of the method, this lack of confidence is mainly related to ambiguous formulations in econometrics, such as the definition of selection bias, selection of core control variables, and method of testing for robustness. Within the framework of the causal models, we clarify the assumption of causal inference using regression-based statistical controls, as described in econometrics, and discuss how to select core control variables to satisfy this assumption and conduct robustness tests for regression estimates.

The relationship between international crude oil prices and China＇s refined oil prices based on a structural VAR model

With the frequent fluctuations of international crude oil prices and China＇s increasing dependence on foreign oil in recent years, the volatility of international oil prices has significantly influenced China domestic refined oil price. This paper aims to investigate the transmission and feedback mechanism between international crude oil prices and China＇s refined oil prices for the time span from January 2011 to November 2015 by using the Granger causality test, vector autoregression model, impulse response function and variance decomposition methods. It is demonstrated that variation of international crude oil prices can cause China domestic refined oil price to change with a weak feedback effect. Moreover, international crude oil prices and China domestic refined oil prices are affected by their lag terms in positive and negative directions in different degrees. Besides, an international crude oil price shock has a signif- icant positive impact on domestic refined oil prices while the impulse response of the international crude oil price variable to the domestic refined oil price shock is negatively insignificant. Furthermore, international crude oil prices and domestic refined oil prices have strong historical inheri- tance. According to the variance decomposition analysis, the international crude oil price is significantly affected by its own disturbance influence, and a domestic refined oil price shock has a slight impact on international crude oil price changes. The domestic refined oil price variance is mainly caused by international crude oil price disturbance, while the domestic refined oil price is slightly affected by its own disturbance. Generally, domestic refined oil prices do not immediately respond to an international crude oil price change, that is, there is a time lag.

基于非稳态加性噪声模型的因果发现算法

因果发现旨在通过观测数据挖掘变量间的因果关系。现有的因果发现方法大多假定数据的产生过程是平稳的,然而在实际环境下往往不满足稳态假设,导致结果不可靠。研究发现,在一些场景中的非稳态扰动与时序信息高度相关。因此,在加性噪声模型基础上将非稳态扰动刻画为一项关于时序信息的函数,设计非稳态加性噪声模型,并给出非稳态加性噪声模型的识别条件,提出一种两阶段的因果关系学习算法。第1阶段利用回归计算得到变量残差,再检验残差与回归特征集的独立性从而选出叶子节点,迭代得到观测变量集的因果次序;第2阶段再次进行回归计算和独立性检验,消除第1阶段中冗余的因果关系,从而得到观测变量集的因果结构。实验结果表明,与基于约束的异构/非平稳因果发现、LPCMCI和Ti MINo算法相比,该算法在仿真数据集上取得了最优的效果,平均F1值达到0.85;而在真实因果结构数据集中,该算法的F1值平均提升41.12%,能够从非稳态数据集中恢复出更多因果结构的信息。

因果关系表示增强的跨领域命名实体识别

跨领域命名实体识别在现实应用中,尤其在目标领域数据稀缺的小样本场景中具有重要价值。然而,现有方法主要是通过特征表示或模型参数共享实现的跨领域实体能力迁移,未充分考虑由于样本选择偏差而引起的虚假相关性问题。为了解决跨领域中的虚假相关性问题,提出一种因果关系表示增强的跨领域命名实体识别模型,将源域的语义特征表示与目标域的语义特征表示进行融合,生成一种增强的上下文语义特征表示。通过结构因果模型捕捉增强后的特征变量与标签之间的因果关系。在目标域中应用因果干预和反事实推断策略,提取存在的直接因果效应,从而进一步缓解特征与标签之间的虚假相关性问题。该方法在公共数据集上进行了实验,实验结果得到了显著提高。

交通安全意识对非机动车骑行者危险骑行行为的影响研究

安全意识在促进安全行为方面发挥着重要作用,但由于安全意识具有多维性和复杂性,难以直接测量。为探究交通安全意识对危险骑行行为的影响,通过云模型选取安全态度、危险认知、安全素质和外界环境这4个潜变量,作为影响交通安全意识的结构要素,并基于调查问卷数据开展实证研究。运用Mplus 8.0软件构建“交通安全意识-危险骑行行为”结构方程模型,量化交通安全意识各要素作用于危险骑行行为的因果链路。采用Bootstrap法检验安全素质、危险认知和安全态度的中介作用,梳理外界环境对危险骑行行为的直接和间接关系;再利用分层回归模型,验证交通安全知识在交通安全意识与危险骑行行为间的调节效应。研究结果表明:①结构方程模型拟合良好,交通安全意识的4个要素分别与危险骑行行为呈显著的负相关,其中,危险认知对无意行为的影响最大(-0.331),安全态度对有意行为的影响最大(-0.332);②中介效应显示外界环境作为外生变量可直接作用于行为,也可通过安全素质、危险认知和安全态度对骑行者的行为产生影响;③交通安全知识的调节作用显著(ΔR^(2)=0.017,P<0.05),该变量强化了交通安全意识与危险骑行行为的负向影响关系,其简单斜率关系表明,当骑行者交通安全知识水平较高时,交通安全意识对危险骑行行为的作用效果更强。

Causalities between Price, Pond Area and Employment in Aquaculture Production

The role of aquaculture industry is becoming more prominent in order to supplement marine capture in meeting the food need for the growing Malaysian population. In an attempt to minimize depletion of marine fisheries, only traditional vessels are allowed to fish along the coastal area while bigger vessels are relegated to deep-sea fishing. During the 9th Malaysian Plan (2006-2010) aquaculture has been recognized as the engine of growth in the national food sector’s development strategy. Future fisheries policy is expected to focus more on aquaculture production, marketing and technological improvement as an alternative to marine capture. This paper investigates the causalities between the selected freshwater fish prices, aquaculture area and production. The study aspires to establish whether or not market price is a key contributor to a rise in the aquaculture area and production. Aquaculture firms comprising the individual culturists are generally motivated by the economic potential of the industry which is reflected in excess of price over cost of production. Our hypothesis is that government policy and initiation rather than prices had give rise to greater participation of culturists and hence augmented the level of employment. However, production increase has a negative implication on environment degradation. Thus there is a conflicting view as regards to the employment opportunity generated by aquaculture undertakings and the need for sustainable development arising from this growing industry. Multivariate time series analysis was used in this investigation.

基于因果正则化极限学习机的风电功率短期预测方法

随着风电并网比例的逐年提高,电力系统对风电功率预测的准确性和稳定性提出了更高要求。对于同一风电场而言,为了避免不同特征选择方法所选择的风电场特征子集不同,从因果关系的角度出发,提出了一种基于因果正则化极限学习机(causal regularized extreme learning machine, CRELM)的风电功率短期预测方法。首先将极限学习机(extreme learning machine, ELM)建模为结构因果模型(structural causal model, SCM),在此基础上计算隐藏层神经元与输出层神经元之间的平均因果效应向量。然后将该平均因果效应向量与输出层权重相结合构成因果正则化项,在最小化训练误差的同时最大化网络的因果关系,以进一步提升模型的预测准确性和预测稳定性。最后,以国内蒙西某风电场数据为例,与采用特征选择或不采用特征选择的预测模型相对比,验证了所提方法的有效性和适用性。

郑州“7·20”地铁水淹事件STAMP致因分析

为探讨河南郑州“7·20”地铁5号线伤亡事件致因,综合运用系统理论事件模型与过程(STAMP)模型,分析5号线伤亡事件安全控制结构,据此逐级辨识事件致因;基于复杂网络理论将事件致因及其联系抽象为网络的节点和边,建立地铁水灾事件网络;计算聚类系数、度数等拓扑参数定量分析事件致因性质,确定关键致因。结果表明:地方党委政府面对汛情的应对部署不紧不实,缺少有效的组织动员和有关部门对工程建设存在失管失察是事件的深层根源;建设单位、设计单位等五方主体单位及运维单位的主体责任失职是事件的主观关键致因;应从政府及有关部门、五方主体单位和运维单位方面采取多元协同的风险防控措施。

Confounding of Three Binary-Variable Counterfactual Model with DAG

Confounding of three binary-variable counterfactual model with directed acyclic graph (DAG) is discussed in this paper. According to the effect between the control variable and the covariate variable, we investigate three causal counterfactual models: the control variable is independent of the covariate variable, the control variable has the effect on the covariate variable and the covariate variable affects the control variable. Using the ancillary information based on conditional independence hypotheses and ignorability, the sufficient conditions to determine whether the covariate variable is an irrelevant factor or whether there is no confounding in each counterfactual model are obtained.

Research on Qualitative Model Decomposition

Qualitative reasoning uses incomplete knowledge to compute a description of the possible behaviors for dynamic systems. A standard qualitative simulation（QSIM） algorithm frequently results in a large number of incomprehensible behavioral descriptions and the simulation for complex systems frequently is intractable. Two model de- composition methods are proposed in this paper to eliminate or decrease the insujficiency of this algorithm. Using a directed graph to represent the qualitative model, the strongly connected graph based theory and genetic algorithm based model decomposition are proposed to decompose the model. A new simple system model is reconstructed by subgraphs and causal relations when the system directed graph is decomposed completely. Each sub-graph is viewed as a separate system and will be simulated separately, and the simulation result of causally upstream subsystem is used to constrain the behavior of downstream subsystems. The model decomposition algorithm provides a promising paradigm for qualitative simulation whose complexity is driven by the complexity of the problem specification rather than the inference mechanism used.