Handling incomplete data in survival analysis with multiple covariates

This paper studies the missing covariate problem which is often encountered in survival analysis. Three covariate imputation methods are employed in the study, and the effectiveness of each method is evaluated within the hazard prediction framework. Data from a typical engineering asset is used in the case study. Covariate values in some time steps are deliberately discarded to generate an incomplete covariate set. It is found that although the mean imputation method is simpler than others for solving missing covariate problems, the results calculated by it can differ largely from the real values of the missing covariates. This study also shows that in general, results obtained from the regression method are more accurate than those of the mean imputation method but at the cost of a higher computational expensive. Gaussian Mixture Model (GMM) method is found to be the most effective method within these three in terms of both computation efficiency and predication accuracy.

Many paths lead to discovery : analogical retrieval of cancer therapies

This paper addresses the issue of analogical inference, and its potential role as the mediator of new therapeutic discoveries, by using disjunction operators based on quantum connectives to combine many potential reasoning pathways into a single search expression. In it, we extend our previous work in which we developed an approach to analogical retrieval using the Predication-based Semantic Indexing (PSI) model, which encodes both concepts and the relationships between them in high-dimensional vector space. As in our previous work, we leverage the ability of PSI to infer predicate pathways connecting two example concepts, in this case comprising of known therapeutic relationships. For example, given that drug x TREATS disease z, we might infer the predicate pathway drug x INTERACTS WITH gene y ASSOCIATED WITH disease z, and use this pathway to search for drugs related to another disease in similar ways. As biological systems tend to be characterized by networks of relationships, we evaluate the ability of quantum-inspired operators to mediate inference and retrieval across multiple relations, by testing the ability of different approaches to recover known therapeutic relationships. In addition, we introduce a novel complex vector based implementation of PSI, based on Plate’s Circular Holographic Reduced Representations, which we utilize for all experiments in addition to the binary vector based approach we have applied in our previous research.

Internet usage trends in medical informatics

Technological growth in the 21st century is exponential. Simultaneously, development of the associated risk, uncertainty and user acceptance are scattered. This required appropriate study to establish people accepting controversial technology (PACT). The Internet and services around it, such as World Wide Web, e-mail, instant messaging and social networking are increasingly becoming important in many aspects of our lives. Information related to medical and personal health sharing using the Internet is controversial and demand validity, usability and acceptance. Whilst literature suggest, Internet enhances patients and physicians’ positive interactions some studies establish opposite of such interaction in particular the associated risk. In recent years Internet has attracted considerable attention as a means to improve health and health care delivery. However, it is not clear how widespread the use of Internet for health care really is or what impact it has on health care utilisation. Estimated impact of Internet usage varies widely from the locations locally and globally. As a result, an estimate (or predication) of Internet use and their effects in Medical Informatics related decision-making is impractical. This open up research issues on validating and accepting Internet usage when designing and developing appropriate policy and processes activities for Medical Informatics, Health Informatics and/or e-Health related protocols. Access and/or availability of data on Internet usage for Medical Informatics related activities are unfeasible. This paper presents a trend analysis of the growth of Internet usage in medical informatics related activities. In order to perform the analysis, data was extracted from ERA (Excellence Research in Australia) ranked “A” and “A*” Journal publications and reports from the authenticated public domain. The study is limited to the analyses of Internet usage trends in United States, Italy, France and Japan. Projected trends and their influence to the field of medical informatics is reviewed and discussed. The study clearly indicates a trend of patients becoming active consumers of health information rather than passive recipients.

A Bayesian regression approach to assess uncertainty in pollutant wash-off modelling

Due to knowledge gaps in relation to urban stormwater quality processes, an in-depth understanding of model uncertainty can enhance decision making. Uncertainty in stormwater quality models can originate from a range of sources such as the complexity of urban rainfall-runoff-stormwater pollutant processes and the paucity of observed data. Unfortunately, studies relating to epistemic uncertainty, which arises from the simplification of reality are limited and often deemed mostly unquantifiable. This paper presents a statistical modelling framework for ascertaining epistemic uncertainty associated with pollutant wash-off under a regression modelling paradigm using Ordinary Least Squares Regression (OLSR) and Weighted Least Squares Regression (WLSR) methods with a Bayesian/Gibbs sampling statistical approach. The study results confirmed that WLSR assuming probability distributed data provides more realistic uncertainty estimates of the observed and predicted wash-off values compared to OLSR modelling. It was also noted that the Bayesian/Gibbs sampling approach is superior compared to the most commonly adopted classical statistical and deterministic approaches commonly used in water quality modelling. The study outcomes confirmed that the predication error associated with wash-off replication is relatively higher due to limited data availability. The uncertainty analysis also highlighted the variability of the wash-off modelling coefficient k as a function of complex physical processes, which is primarily influenced by surface characteristics and rainfall intensity.

The non-literary Latin letters. A study of their syntax and pragmatics

This dissertation studies the language of Latin letters that were written in Egypt and Vindolanda (in northern Britain) during the period 1st century BC 3rd century AD on papyri, ostraca, and wooden tablets. The majority of the texts is, in one way or another, connected with the Roman army. The focus of the study is on syntax and pragmatics. Besides traditional philological methods, modern syntactic theory is used as well, especially in the pragmatic analysis. The study begins with a critical survey of certain concepts that are current in the research on the Latin language, most importantly the concept of vulgar Latin , which, it is argued, seems to be used as an abstract noun for variation and change in Latin . Further, it is necessary to treat even the non-literary material primarily as written texts and not as straightforward reflections of spoken language. An examination of letter phraseology shows that there is considerable variation between the two major geographical areas of provenance. Latin letter writing in Egypt was influenced by Greek. The study highlights the importance of seeing the letters as a text type, with recurring phraseological elements appearing in the body text as well. It is argued that recognising these elements is essential for the correct analysis of the syntax. Three areas of syntax are discussed in detail: sentence connection (mainly parataxis), syntactically incoherent structures and word order (the order of the object and the verb). For certain types of sentence connection we may plausibly posit an origin in spoken Latin, but for many other linguistic phenomena attested in this material the issue of spoken Latin is anything but simple. Concerning the study of historical syntax, the letters offer information about the changing status of the accusative case. Incoherent structures may reflect contaminations in spoken language but usually the reason for them is the inability of the writer to put his thoughts into writing, especially when there is something more complicated to be expressed. Many incoherent expressions reflect the need to start the predication with a thematic constituent. Latin word order is seen as resulting from an interaction of syntactic and pragmatic factors. The preference for an order where the topic is placed sentence-initially can be seen in word order more generally as well. Furthermore, there appears a difference between Egypt and Vindolanda. The letters from Vindolanda show the order O(bject) V(erb) clearly more often than the letters from Egypt. Interestingly, this difference correlates with another, namely the use of the anaphoric pronoun is. This is an interesting observation in view of the fact that both of these are traditional Latin features, as opposed to those that foreshadow the Romance development (VO order and use of the anaphoric ille). However, it is difficult to say whether this is an indication of social or regional variation.

Water Content Distribution in the Surface Layer of Maoping Slope

The water content distribution in the surface layer of Maoping slope has been studied by testing the water content at 31 control sites. The water content profiles at these sites have also been determined. The water content distributions at different segments have been obtained by using the Kriging method of geostatistics. By comparing the water content distributions with the landform of the slope, it was shown that the water content is closely dependent on the landform of the slope. The water content distribution in the surface layer provided a fundamental basis for landslide predication and treatment.

A transitividade verbal: uma revisão semântico-pragmática

Esta tese aborda o tema Transitividade Verbal em construções predicativas encontradas em corpus construído com textos jornalísticos impressos do Brasil. Os verbos são estudados segundo ocorrências retiradas de jornais de grande circulação em três capitais. Toma-se a semântica, a pragmática e a sintaxe como determinantes potenciais da transitividade verbal e aborda-se o tema segundo perspectiva histórica, ou seja, procura-se desenhar o percurso do tema nos estudos de língua portuguesa desde a sua primeira gramática, chegando-se, então, a reflexões mais atuais e também ao tratamento dispensado aos verbos e sua construção predicativa em dicionários e em gramáticas pedagógicas. O confronto com estudos acerca da língua espanhola também é motivo de atenção, devido à proximidade que o tema apresenta nas duas línguas. A partir desses estudos, buscou-se elaborar quadro sistemático e mais coerente dos verbos segundo a transitividade e construção predicativa, procurando aliar-se a base teórica consistente, apresentada pela tradição dos estudos gramaticais, às reflexões técnico-científicas acordadas com a realidade lingüística levantada a partir do corpus

Os impasses da sublimação: articulações entre psicanálise e estética

A sublimação é um conceito forjado por Freud que tem por finalidade responder sobre a incidência da sexualidade num campo que não é diretamente ligado à função sexual. Freud inicialmente ligou o conceito de sublimação ao desenvolvimento da sociedade, mantendo entre eles uma estreita ligação através da idéia de aprovação social, coerência e harmonia, sendo a sexualidade algo que deveria ser posta a serviço da civilização pela educação e restrição de sua manifestação. A importância dada aos aspectos quantitativos presentes na subjetividade fez com que a sublimação adquirisse maior destaque, sendo diretamente ligada ao destino a ser dado aos elementos que resistiam à lógica da predicação e significação, buscando alternativas para a manifestação desses fatores que ameaçavam constantemente desestruturar o psiquismo. A sublimação se torna um processo necessário para articular a subjetividade a uma lógica da inadequação e do inacabamento, mostrando-se um operador não só da clínica como também da cultura, pois leva em consideração a sustentação da posição desejante sem incremento de mal-estar. Esses fatores aproximam de maneira surpreendente estética e psicanálise através de um novo campo de atuação. Freud percebeu a capacidade terapêutica da sublimação no final de sua vida, e Lacan, outro psicanalista que se mostrou muito interessado neste tema, também nos deixou uma série de elaborações que podem nos ser úteis para enfrentar os desafios que o nosso tempo produz.

Photoluminescence of low-dimensional semiconductor structures under pressure

Photoluminescence of some low-dimensional semiconductor structures has been investigated under pressure. The measured pressure coefficients of In0.55Al0.45 As/Al0.5Ga0.5As quantum dots with average diameter of 26, 52 and 62 nm are 82, 94 and 98 meV/GPa, respectively. It indicates that these quantum dots are type-I dots. On the other hand, the measured pressure coefficient for quantum dots with 7 nm in size is -17meV/GPa, indicating the type-II character. The measured pressure coefficient for Mn emission in ZnS:Mn nanoparticles is -34.6meV/GPa, in agreement with the predication of the crystal field theory. However, the DA emission is nearly independent on pressure, indicating that this emission is related to the surface defects in ZnS host. The measured pressure coefficient of Cu emission in ZnS: Cu nanoparticles is 63.2 meV/GPa. It implies that the acceptor level introduced by Cu ions has some character of shallow level. The measured pressure coefficient of Eu emission in ZnS:Eu nanoparticles is 24.1 mev/GPa, in contrast to the predication of the crystal field theory. It may be due to the strong interaction between the excited state of Eu ions and the conduction band of ZnS host.

一种基于谓词规则的复杂文字处理模型

复杂文字在显示输出的过程中,表现出极为复杂的语言特征.为此提出了一种基于谓词规则的复杂文字处理模型,模型以谓词规则的方法给出了复杂文字字形布局特征的形式化描述,按照复杂文字处理的过程,设计了实现该模型的软件体系结构,将复杂文字的语言特征从程序控制逻辑中隔离出来,提高了系统的灵活性,便于增加新的复杂文字的支持.在研制蒙古文、藏文、维吾尔文办公套件的应用中表明,该模型是实用有效的.

基于方面模板的分布式组件系统性能预测方法

目前分布式组件系统通常需要中间件提供的横切关注点(crosscutting)实现非功能性特征,而这种设计方法在简化分布式组件系统开发过程的同时难以在设计时进行有效的性能预测.研究了一种基于方面模板的分布式组件系统性能预测方法,将横切关注点提炼为可复用的方面模板,并通过面向方面的建模技术,自动构建包括中间件横切关注点和中间件性能因素的完整组件性能模型.该模型的预测结果可辅助设计人员尽早地发现组件设计缺陷或帮助筛选备选方案.

Study on stimulated emission of Eu2+ in crystal KMgF3 : Eu2+

The line emission peak of Eu2+ ion in crystal KMgF3 is at 360 nm. The probability of stimulated emission on 4f(7) (P-6(7/2))-> 4f(7)(S-8(7/2)) transition was predicted with a four-level decay model of Eu2+6P7/2 excited states proposed by the authors. Optic gain and net gain coefficient (g = 11.4 +/- 3.2)cm(-1) of 360 nm emission in crystal KMgF3 : Eu2+ were measured by ASE method, and the predication was proved by experiment. The net gain coefficient can be increased by annealing or doping crystal KMgF3 : Eu2+ with Gd3+ or Ce3+.

鄂尔多斯盆地上古生界低渗透砂岩有效储层识别与预测

Sulige Gasfield, with a basically proven reserve as high as one trillion cubic meters, is one giant gas field discovered in China. The major gas -bearing layers are Upper Paleozoic strata with fluvial-lacustrine sedimentary facies. Generally, gas reservoirs in this field are characteristic by "five low" properties, namely low porosity, low permeability, low formation pressure, low productivity and low gas abundance. Reservoirs in this field also feature in a large distribution area, thin single sandbody thickness, poor reservoir physical properties, thin effective reservoir thickness, sharp horizontal and/or vertical changes in reservoir properties as well as poor connectivity between different reservoirs. Although outstanding achievements have been acquired in this field, there are still several problems in the evaluation and development of the reservoirs, such as: the relation between seismic attributes and reservoir property parameters is not exclusive, which yields more than one solution in using seismic attributes to predict reservoir parameters; the wave impedance distribution ranges of sandstone and mudstone are overlapped, means it is impossible to distinguish them through the application of post-stack impedance inversion; studies on seismic petrophysics, reservoir geophysical properties, wave reflection models and AVO features have a poor foundation, makes it difficult to recognize the specific differences between tight sandstone and gas-bearing sandstone and their distribution laws. These are the main reasons causing the low well drilling success rate and poor economic returns, which usually result in ineffective development and utilization of the field. Therefore, it is of great importance to perform studies on identification and prediction of effective reservoirs in low permeable sandstone strata. Taking the 2D and 3D multiwave-multicomponent seismic exploration block in Su6-Su5 area of Sulige field as a study area and He 8 member as target bed, analysis of the target bed sedimentary characteristics and logging data properties are performed, while criteria to identify effective reservoirs are determined. Then, techniques and technologies such as pre-stack seismic information (AVO, elastic impedance, wave-let absorption attenuation) and Gamma inversion, reservoir litological and geophysical properties prediction are used to increase the precision in identifying and predicting effective reservoirs; while P-wave and S-wave impedance, ratio of P/S wave velocities, rock elastic parameters and elastic impedance are used to perform sandstone gas-bearing property identification and gas reservoir thickness prediction. Innovative achievements are summarized as follows: 1. The study of this thesis is the first time that multiwave-multicomponent seismic data are used to identify and predict non-marine classic reservoirs in China. Through the application of multiwave-multicomponents seismic data and integration of both pre-stack and post-stack seismic data, a set of workflows and methods to perform high-precision prediction of effective reservoirs in low permeable sandstone is established systematically. 2. Four key techniques to perform effective reservoir prediction including AVO analysis, pre-stack elastic wave impedance inversion, elastic parameters inversion, and absorption attenuation analysis are developed, utilizing pre-stack seismic data to the utmost and increasing the correct rate for effective reservoir prediction to 83% from the former 67% with routine methods. 3. This thesis summarizes techniques and technologies used in the identification reservoir gas-bearing properties using multiwave-multicomponent seismic data. And for the first time, quantitative analysis on reservoir fluids such as oil, gas, and/or water are carried out, and characteristic lithology prediction techniques through the integration of pre-stack and post-stack seismic prediction techniques, common seismic inversion and rock elastic parameters inversion, as well as P-wave inversion and converted wave inversion is put forward, further increasing the correct rate of effective reservoir prediction in this area to 90%. 4. Ten seismic attribute parameters are selected in the 3D multi-wave area to perform a comprehensive evaluation on effective reservoirs using weighted-factor method. The results show that the first class effective reservoir covers an area of 10.08% of the study area, while the second and the third class reservoirs take 43.8% and 46% respectively, sharply increasing the success rate for appraisal and development wells.