基于多源数据融合的复杂疾病建模方法研究

复杂疾病(如癌症)严重危害人类身体健康,其发生和发展往往是多因素共同作用的结果,如何清楚地认识疾病的内在致病机理是当前研究的重点内容.目前,随着高通量技术的发展,大量的生物数据涌现,因此,以数据为驱动,通过对生物系统的多源数据融合来实现复杂疾病的建模成为了一个新的研究热点.由于生物系统的多源数据往往在数据分布、尺度和格式等方面存在巨大差异,这使得如何有效地融合数据成为建模过程中的重点.本文从多源数据融合的角度,分别对疾病亚型识别、疾病相关模块挖掘和样本特异性疾病相关标志物识别3个方面对人类复杂疾病的建模方法进行讨论,力图为研究人员在人类复杂疾病的数据建模方面提供建议.     

医疗质量综合评价指标可比性研究

应用直接标化法计算病种构成变化对医疗质量综合评价指标的影响程度并算出标化值。１９９１年病种构成变化对治愈率、病死率、平均住院日变化的影响程度分别为６９．９１％，８４．９６％，８６．７７％，表明病种构成变化对医疗质量综合评价指标的影响不容忽视，未经标化的指标无可比性     

住院患者病情危重度的分类决策树研究

目的　应用分类决策树对各类疾病住院患者的病情进行分类和评价。方法　采用SPSSAnswerTree分类软件和CHAID算法,用1244887例住院患者的病案首页数据,通过确定分类目标变量、预测变量、决策树的生长和修剪规则,建立覆盖所有病种的危重度分类模型。结果　建立了163个基于病种的分类模型,每个病种都形成了若干个危重度等级,共产生了636个危重度类别,每个类别对应一个危重度分值。经新样本考核,该病情指标对治疗结果、医院规模有一定预测能力,且与住院费用相关。结论　病情分类结果能够反映住院患者的相对病情,可作为病情标准化的依据,用于医疗服务质量效率评价。     

高血压病的风险因素分析与研究

高血压病是全球性多发病之一，其发病率在中国逐年上升，影响高血压疾病的因素很多，本文研究的是年龄、BMI、家族史、吸烟时间、吸烟数量、饮酒时间、饮酒频率、饮酒数量和户外活动，共九个因素，研究的主要目的是将这些因素对高血压患病影响的重要性程度进行排序，并建立相应患病概率的预测模型，这不仅对医疗保险中保单的核保提供了风险度量的判别依据，而且对于人们日常生活疾病的预防、监视也有一定的指导意义。本文首先用单因子Logistic回归剔除与高血压患病相关性较小的因素，继而用主成分分析方法消除因素的共线性，最后用Logistic回归拟合患病概率模型并根据系数进行排序和预测。结果无论是男性还是女性，对高血压病影响最大的都是家族史，其次为BMI和年龄。     

妊娠期高血压疾病危险因素的统计分析

妊娠期高血压疾病(以下简称妊高病)是妊娠期特有的疾病.随着生活节奏快、精神压力大、高龄初产妇增多等高危因素的凸现,妊高病倾向者也逐渐增多.目前,对妊高病发病高危因素的研究很多.本文基于双重logistic回归模型对影响妊高病的危险因素进行变量选择和预测分析.     

桥本甲状腺炎合并甲状腺癌的危险因素回归分析

目的 探讨桥本甲状腺炎（HT）合并甲状腺癌的危险因素。方法 采用回顾性病例对照研究,选取59 例HT 合并甲状腺癌患者为病例组,按照1∶2比例选取118 例HT 合并甲状腺良性结节或单纯HT 患者为对照组,对相关因素进行单因素和多因素logistic 回归分析。结果 单因素分析显示甲状腺疾病家族史、碘摄入情况、辐射接触史、甲状腺自身抗体和合并单发结节为HT 合并甲状腺癌的影响因素。经多因素分析家族史、高碘摄入、辐射接触史为HT 合并甲状腺癌的危险因素,其OR 值（95%CI）分别为2.141（1.664~2.755）、2.479（1.895~2.936）和4.596（3.693~4.997）。结论 HT 合并甲状腺癌发病机制有待于进一步研究,应针对危险因素(家族史、碘摄入情况、辐射接触史)采取措施进行早防早治。     

代谢综合症的患病因素分析与预测

以高血压患者的患病因素为例进行解析,提出代谢综合症相关问题的解决方法.研究年龄、BMI、家族史、吸烟、饮酒、文化程度、职业等18个影响高血压疾病的因素.研究的主要目的是将这些因素对高血压患病影响的重要性程度进行排序,并建立相应患病概率的预测模型.这不仅对人们日常生活疾病的预防、监控有一定的指导意义,也给医疗保险中保单的核保提供了风险度量的判别依据.首先用相关性分析剔除与高血压患病相关性较小的因素,继而用主成分分析方法消除因素间的共线性,最后用Logistic回归拟合患病概率模型并根据系数进行排序和预测.结果显示,无论性别,对高血压疾病影响最大的首先是家族史,其次为年龄.     

具有遗传性疾病和性状的遗传位点分析

随着信息技术的进步和发展,现代生物学越来越多地将这些技术用于大规模生物数据的收集、分析、挖掘等过程.大量计算机技术,特别是统计方法被用来进行复杂疾病的分析.大量研究表明,人体的许多表型性状差异以及对药物和疾病的易感性等都可能与某些位点相关联,或和包含有多个位点的基因相关联.因此,定位与性状或疾病相关联的位点在染色体或基因中的位置,能帮助研究人员了解性状和一些疾病的遗传机理,也能使人们对致病位点加以干预,防止一些遗传病的发生.利用随机森林方法、Bootstrap重抽样、logistic回归等大数据分析方法,意在解决优化生物学位点关联性分析中单一致病位点识别、多位点相互作用和多性状位点关联性分析等子问题.     

多风险因素的投标报价决策方法

本文讨论工程项目投标报价的多风险因素层次模型，系统地介绍了90年代以来发展出的多风险因素条件下投标报价的5种主要决策方法，即层次分析法（AHP）、人工神经网络（ANN）、模糊评价法（Fuzzy）、专家系统（ES）和基于事例推理（CBR）。本文也对中国在该领域的研究现状作一个简单的评述。     

因素交互作用分析的广义线性模型

为了估计疾病暴露因素之间的交互作用,提出了一种分析疾病暴露因素的交互作用的有效方法.该方法建立一个广义线性模型,通过估计模型中的参数及方差、协方差,定量的分析交互作用的大小和类型,不仅适用于队列研究资料,同时也适用用于病例-对照研究资料.     

股票市盈率水平的决定因素——上海证券交易所的数据验证

本文利用截面分析方法,从多个角度讨论影响股票市盈率水平的因素,并利用上海证券交易所股票数据进行验证。首先,利用上海证券交易所的股票历史数据计算按不同板块 (如行业,每股收益大小等)划分股票类别时,各板块市盈率水平的具体数值,并讨论各不同板块市盈率的分布特征。其次,讨论了市盈率与其直接影响变量(股票价格和每股收益 )之间的相互关系,用数学模型论述两者在影响市盈率变化时所存在的差异。最后,通过对影响市盈率水平的各种因素进行详细的统计特征分析,筛选影响强烈的因素,并建立了描述市盈率与各影响因素之间相关性的多因素模型。     

Choice of modelling technique for evaluating health care interventions

Economic evaluation, such as cost effectiveness analysis, provides a method for comparing healthcare interventions. These evaluations often use modelling techniques such as decision trees, Markov processes and discrete event simulations (DES). With the aid of examples from coronary heart disease, the use of these techniques in different health care situations is discussed. Guidelines for the choice of modelling technique are developed according to the characteristics of the health care intervention.The choice of modelling technique is shown to depend on the acceptance of the modelling technique, model ‘error’, model appropriateness, dimensionality and ease and speed of model development. Generally decision trees are suitable for acute interventions but they cannot model recursion and Markov models are suitable for simple chronic interventions. It is further recommended that population based models be used in order to provide health care outcomes for the likely cost, health benefits and cost effectiveness of the intervention. The population approach will complicate the construction of the model. DES will allow the modeller to construct more complex, dynamic and accurate systems but these may involve a corresponding increase in development time and expense. The modeller will need to make a judgement on the necessary complexity of the model in terms of interaction of individuals and model size and whether queuing for resources, resource constraints or the interactions between individuals are significant issues in the health care system.     

Output learning and duality in joint production programs

Summary Output learning is incorporated into a short-run static cost-minimizing model of the multiproduct, multifactor firm which employs a fixed-coefficients technology. The firm's output processes or activities are ultimately specified as functions of the activity variables themselves, thus rentering a generalization to a concave program. A Lagrange dual formulation is then used to obtain the indirect cost objective. Given that this optimal cost function is differentiable and satisfies a regularity condition, its price derivatives serve as input demand functions while its derivatives with respect to the minimum output requirements yield a set of (implicit) marginal costs or dual variables.     

基于CSO-SVR的气膜钢筋混凝土储仓主体结构施工成本预测研究

为了较准确的预测气膜钢筋混凝土储仓主体结构施工成本,提出一种鸡群算法(CSO)和支持向量回归机(SVR)结合模型,即CSO-SVR,利用CSO算法对SVR进行寻优得到全局最优解,从而得到具有最佳参数的支持向量回归机模型,通过气膜钢筋混凝土储仓主体结构施工成本数据预测仿真,结果显示:CSO-SVR模型预测精度高于PSO-SVR,GA-SVR,SVR,BPNN等方法,是预测气膜钢筋混凝土储仓主体结构施工成本的有效工具.     

Economic implications of poor access to antenatal care in rural and remote Western Australian Aboriginal communities: An individual sampling model of pregnancy

Background

Australian Aboriginal women attend antenatal care less frequently and experience poorer pregnancy outcomes than non-Aboriginal women. Improving access to antenatal care is recognised as a means to improve pregnancy outcomes for mother and baby.

Objective

To estimate the costs of inadequate antenatal care and provide baseline measurements and implications for policy that targets improving access to care in rural and remote Western Australian (WA) Aboriginal communities.

Methods

An individual sampling model of pregnancy was developed that simulated hypothetical women with pregnancy events and outcomes observed in the WA Aboriginal population. Weekly pregnancy events were modelled via logistic regression according to maternal characteristics, events during pregnancy and current gestation, with adequate and inadequate care (?4 and <4 antenatal visits) to reflect differences in outcomes reported in the literature. The pregnancy model simulated clinical management including antenatal visits, hospitalisations, and transfers to tertiary care.

Results

The mean cost of pregnancy was AUD$8985, with a large difference depending on access to antenatal care (AUD$7635 and AUD$10,216 for adequate and inadequate care respectively). The main difference in costs resulted from neonatal care (AUD$1021 vs AUD$3205 for adequate and inadequate care respectively). In a rural community with 150 births per year, up to AUD$123,082 may be spent to improve access to care at no extra cost to the total current expenditure (AUD$1,347,733).

Conclusions

The large difference in pregnancy costs between those receiving adequate and inadequate care demonstrates that additional expenditure on improving access to antenatal care may be cost-effective and should be further investigated.     

Using phase-type models to cost stroke patient care across health,social and community services

Stroke disease places a heavy burden on society, incurring long periods of time in hospital and community care, and associated costs. Also stroke is a highly complex disease with diverse outcomes and multiple strategies for therapy and care. Previously a modeling framework has been developed which clusters patients into classes with respect to their length of stay (LOS) in hospital. Phase-type models were then used to describe patient flows for each cluster. Also multiple outcomes, such as discharge to normal residence, nursing home, or death can be permitted. We here add costs to this model and obtain the Moment Generating Function for the total cost of a system consisting of multiple transient phase-type classes with multiple absorbing states. This system represents different classes of patients in different hospital and community services states. Based on stroke patients’ data from the Belfast City Hospital, various scenarios are explored with a focus on comparing the cost of thrombolysis treatment under different regimes. The overall modeling framework characterizes the behavior of stroke patient populations, with a focus on integrated system-wide costing and planning, encompassing hospital and community services. Within this general framework we have developed models which take account of patient heterogeneity and multiple care options. Such complex strategies depend crucially on developing a deep engagement with the health care professionals and underpinning the models with detailed patient-specific data.     

Cost efficiency and scale economies in general dental practices in the US: a non-parametric and parametric analysis of Colorado data

This paper uses both the non-parametric method of data envelopment analysis (DEA) and the econometric method of stochastic frontier analysis (SFA) to study the production technology and cost efficiency of the US dental care industry using practice level data. The American Dental Association 2006 survey data for a number of general dental practices in the state of Colorado in the US are used for the empirical analysis. The findings suggest that the cost efficiency score is between 0.79 and 0.87, on average, and the cost inefficiency is mostly due to allocative rather than technical inefficiency. The optimal output level for a dental practice to fully exploit the economies of scale is estimated to be at $1.68 million. Average cost at this level of output is 50.6 cents for each dollar of gross billing generated. The DEA and SFA approaches provide generally consistent results.     

Optimization of financial and energy structure of productive capital

^** Email: nahritonenko{at}pvamu.edu^*** Email: yyatsenko{at}hbu.edu Optimal control of special non-linear Volterra integral equationsis used to optimize the structure and lifetime of age-dependentproductive capital at an individual enterprise level. The equationsdescribe a multifactor vintage capital model. The optimizationproblem is to select investment, specific capital cost, specificenergy consumption and capital lifetime that maximize net profiton finite and infinite horizons. The structure of optimal trajectoriesis investigated and relevant relations among technological andfinancial parameters of the productive capital are established.     

On the Cost of Data Analysis

Abstract

A regression analysis usually consists of several stages, such as variable selection, transformation and residual diagnosis. Inference is often made from the selected model without regard to the model selection methods that preceeded it. This can result in overoptimistic and biased inferences. We first characterize data-analytic actions as functions acting on regression models. We investigate the extent of the problem and test bootstrap, jackknife, and sample-splitting methods for ameliorating it. We also demonstrate an interactive LISP-STAT system for assessing the cost of the data analysis while it is taking place.     

成长性差异、信息披露与权益资本成本

本文研究成长性差异是否以及如何影响信息披露与权益资本成本之间的关系。实证结果表明成长性差异可以显著调节信息披露与权益资本成本之间的关系,利用分组回归分析法发现,高成长性上市公司信息披露水平越高,权益资本成本越低。进一步研究发现,在高成长性样本组中,非国有控股上市公司信息披露对权益资本成本的影响相对显著,而在低成长性样本组中,国有控股上市公司信息披露对权益资本成本的影响相对显著。在采用工具变量法和两阶段最小二乘法等(2SLS)进行一系列稳健性检验之后,上述结论仍然成立。本文的研究结论为上市公司和监管机构完善信息披露制度提供一定的启示。