Conclusions. The sensitivity and positive predictive value are generally more concerned than the specificity and negative predictive value in assessing and adjusting publication bias in meta-analyses. In this sense, Egger's regression can be recommended for its high sensitivity, while any positive result from Tang's method would suggest a probability of bias that should be taken seriously. Given the different patterns of the accuracy with the OR and the P1-P2 combination, a combination of Egger's regression and Tang's regression would be advisable. Further studies are needed to study the accuracy of methods used in combination. / Due to sampling error and true heterogeneity, a single study cannot provide a comprehensive picture and a precise estimate of, say the effectiveness of a treatment. Systematic reviews that identify and integrate relevant studies have become the most important scientific, quantitative method to summarize scientific research. Meta-analysis is the statistical method used in systematic reviews to combine results from individual studies. / However, due to selective submission and publication, not all relevant studies conducted, especially those unpublished studies with an insignificant negative result, are easily accessible to those who conduct reviews. As a result, the truth, say, the effect of a treatment, would be overestimated. This phenomenon is known as publication bias. A few methods for detecting the bias have been developed and used in meta-analyses. Although their accuracy has been studied, some important issues remain to be answered, such as when would a method be good enough for practical use and is it similarly good for different definitions of the odds ratio? / Methods. We conducted a simulation study to examine the accuracy of four commonly used bias-detection methods with various ORs and P1-P2 combinations. In a simulation study, the true bias status can be predetermined and thus be compared with the results of the bias-detection methods. The four methods are Egger's regression, funnel plot regression, rank correlation regression, and Tang's regression. Realistic sample size was used for simulating individual studies and the numbers of studies in a meta-analysis was also varied. Both the sensitivity and specificity are examined against the magnitude of the OR and the P1-P 2 combination to identify the ORs and P1-P 2 combinations for which a method is sufficiently accurate. Predictive values are also examined for the same reason and in the same manner. / Results. The sensitivity and positive predictive value are generally low and in particular when the OR is close to one for which publication bias is of a particular concern. Egger's regression has the highest sensitivity among the four, in particular when the OR is neither close to one nor exceptionally large or small. Due to the relatively lower specificity, the positive predictive value of Egger's regression is not as high as that for Tang's regression and funnel plot regression. Tang's regression and funnel plot regression are very similar in sensitivity, specificity and predictive values, with the former being slightly better. Rank correlation seems the least accurate method overall. Tang's regression has in general the highest positive predictive value among the four methods in particular when the OR is below one. / Chung Chi-keung. / "June 2006." / Adviser: Tang Jin Ling. / Source: Dissertation Abstracts International, Volume: 68-03, Section: B, page: 1588. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2006. / Includes bibliographical references (p. 116-124). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Religious belief and attendance predict improved well-being at the individual level. Paradoxically, geographic locations with high rates of religious belief and attendance are often those with the differentially high rates of societal instability and suffering. Many of the consequences of religiosity are context-based and vary across time, and holy days are naturally-occurring religious cues that have been shown to influence religiously-relevant attitudes and behaviors. I investigated the degree to which personal religiosity and religious geography (i.e. religious demographics with other location variables) individually and interactively predict well-being across days of the week. In the first study, American Christians demonstrated greater well-being by expressing more positive sentiment in Twitter posts, while American Muslims displayed less well-being. Sundays were generally the most positive day, but American Muslims communicated more happiness on Fridays (the Muslim holy day). In the second study, Christianity did not predict increased well-being in the posts of college students. In the third study, global survey data with measures of religiosity and well-being indicated that the well-being consequences of religious affiliation depend on the religious group and location, and that people tend to be especially positive on their group’s holy day. Study four explored the latent topical content of Twitter posts. Across studies, religious minority status appeared to have a deleterious effect on well-being.

Recentemente, houve um aumento rápido na criação e disponibilidade de repositórios de dados, o que foi percebido nas áreas de Mineração de Dados e Aprendizagem de Máquina. Este fato deve-se principalmente à rápida criação de tais dados em redes sociais. Uma grande parte destes dados é feita de texto, e a informação armazenada neles pode descrever desde perfis de usuários a temas comuns em documentos como política, esportes e ciência, informação bastante útil para várias aplicações. Como muitos destes dados são criados em fluxos, é desejável a criação de algoritmos com capacidade de atuar em grande escala e também de forma on-line, já que tarefas como organização e exploração de grandes coleções de dados seriam beneficiadas por eles. Nesta dissertação um modelo probabilístico, on-line e incremental é apresentado, como um esforço em resolver o problema apresentado. O algoritmo possui o nome DV-INBC e é uma extensão ao algoritmo INBC. As duas principais características do DV-INBC são: a necessidade de apenas uma iteração pelos dados de treino para criar um modelo que os represente; não é necessário saber o vocabulário dos dados a priori. Logo, pouco conhecimento sobre o fluxo de dados é necessário. Para avaliar a performance do algoritmo, são apresentados testes usando datasets populares. / Recently the fields of Data Mining and Machine Learning have seen a rapid increase in the creation and availability of data repositories. This is mainly due to its rapid creation in social networks. Also, a large part of those data is made of text documents. The information stored in such texts can range from a description of a user profile to common textual topics such as politics, sports and science, information very useful for many applications. Besides, since many of this data are created in streams, scalable and on-line algorithms are desired, because tasks like organization and exploration of large document collections would be benefited by them. In this thesis an incremental, on-line and probabilistic model for document classification is presented, as an effort of tackling this problem. The algorithm is called DV-INBC and is an extension to the INBC algorithm. The two main characteristics of DV-INBC are: only a single scan over the data is necessary to create a model of it; the data vocabulary need not to be known a priori. Therefore, little knowledge about the data stream is needed. To assess its performance, tests using well known datasets are presented.

Medical Information Visualization makes optimized use of digitized data of medical records, e.g. Electronic Medical Record. This thesis is an extended work of Pictorial Information Visualization System (PIVS) developed by Yongji Jin (Jin, 2016) Jiaren Suo (Suo, 2017) which is a graphical visualization system by picturizing patient’s medical history summary depicting patients’ medical information in order to help patients and doctors to easily capture patients’ past and present conditions. The summary information has been manually entered into the interface where the information can be taken from clinical notes. This study proposes a methodology of automatically extracting medical information from patients’ clinical notes by using the techniques of Natural Language Processing in order to produce medical history summarization from past medical records. We develop a Named Entities Recognition system to extract the information of the medical imaging procedure (performance date, human body location, imaging results and so on) and medications (medication names, frequency and quantities) by applying the model of conditional random fields with three main features and others: word-based, part-of-speech, Metamap semantic features. Adding Metamap semantic features is a novel idea which raised the accuracy compared to previous studies. Our evaluation shows that our model has higher accuracy than others on medication extraction as a case study. For enhancing the accuracy of entities extraction, we also propose a methodology of Topic Segmentation to clinical notes using boundary detection by determining the difference of classification probabilities of subsequence sequences, which is different from the traditional Topic Segmentation approaches such as TextTiling, TopicTiling and Beeferman Statistical Model. With Topic Segmentation combined for Named Entities Extraction, we observed higher accuracy for medication extraction compared to the case without the segmentation. Finally, we also present a prototype of integrating our information extraction system with PIVS by simply building the database of interface coordinates and the terms of human body parts.

Perdas gestacionais podem ser determinadas por vários fatores, dentre eles a endometriose, doença ginecológica comum caracterizada pela presença e crescimento de glândulas e estroma endometrial fora da cavidade uterina. Os principais sintomas da doença envolvem, além de problemas de fertilidade, dismenorréia, dispaneuria, dor pélvica crônica e irregularidades menstruais. Os locais mais freqüentes das lesões endometrióticas são o peritônio e os órgãos pélvicos, principalmente ovários. A incidência da endometriose é difícil de ser determinada devido à grande variabilidade de sintomas e à dificuldade para confirmação diagnóstica, que requer método cirúrgico. Estima-se que 15% da população feminina em idade reprodutiva seja afetada pela doença, sendo sua patogênese pouco conhecida. A hipótese mais aceita é a da menstruação retrógrada, onde os fragmentos endometriais descamados durante a fase menstrual seriam transportados através das tubas uterinas até a cavidade peritonial com implantação, crescimento e invasão local e de órgãos adjacentes. No entanto, apenas o refluxo não é suficiente para o estabelecimento da doença, sendo necessário que as células endometriais possuam certas características moleculares que favoreçam o aparecimento e a progressão da implantação ectópica. Vários estudos na literatura evidenciam que as principais divergências moleculares entre portadoras e não portadoras de endometriose estão relacionadas a processos envolvidos na apoptose, adesão celular, angiogênese, biosíntese de estrógeno, sistema imune, além de fatores de crescimento e metaloproteinases. Sendo assim, as pesquisas buscam a investigação de genes que se expressem diferencialmente (maior ou menor expressão) nas células de lesões endometrióticas por meio de várias técnicas. A hibridação subtrativa é uma metodologia de screening gênico que compara dois grupos celulares distintos, permitindo isolar moléculas de cDNA representantes do genoma de somente um dos dois grupos, pois remove as seqüências comuns entre eles. A técnica de hibridação subtrativa rápida (RaSH) simplifica e torna mais eficiente esse processo de subtração, identificando grande quantidade de seqüências diferencialmente expressas. Com objetivo de determinar possíveis marcadores moleculares para diagnóstico e tratamento da doença, aplicamos a técnica de RaSH para identificação de genes com expressão diferencial em 11 amostras de lesões endometrióticas (cinco de origem ovariana e seis de origem peritonial) e em 11 amostras de tecido endometrial tópico de mulheres sem a doença. Após análise dos dados referentes a 166 sequências de referência, os genes HTRA1, LOXL1, SPARC e SSAT foram selecionados para validação por meio da técnica de RT-PCR quantitativa em tempo real, sendo que apenas os dois primeiros mostraram diferença significativa entre os dois grupos estudados. Os genes HTRA1 e LOXL1 apresentaram expressão aumentada em lesões endometrióticas de pacientes afetadas, quando comparada aos níveis de expressão no tecido endometrial de mulheres não afetadas. Os resultados obtidos sugerem que os genes HTRA1 e LOXL1 podem ser considerados candidatos a marcadores moleculares para o diagnóstico. / Gestational losses can be determined by a number of factors including endometriosis, a common gynecological disease characterized by endometrial tissue found outside the uterine cavity. The main symptoms of the disease involve dysmenorrhea, dyspaneuria, chronic pelvic pain and menstrual irregularities, besides fertility problems. Endometriotic lesions are more frequently found in the peritoneum and in pelvic organs, mainly ovaries. Endometriosis incidence is difficult to be determined due to the wide variability of the symptoms and the difficult diagnostic confirmation, which requires a surgical method. Nevertheless, it is believed that around 15% of the female population in reproductive age is affected by the disease, although its pathogenesis remains unclear. The most accepted hypothesis is the retrograde menstruation, where endometrial fragments from the menstrual phase are transported through the uterine tubes to the peritoneal cavity, where they undergo implantation, growth, and adjacent tissues invasion. However, only reflux is not enough for the disease establishment, and it is necessary that endometrial cells present molecular characteristics favoring rise and progression of ectopic implantation. In the literature, a number of studies highlight that the main molecular divergences between women with and without endometriosis are related to processes involved in apoptosis, cellular adhesion, angiogenesis, estrogen biosynthesis, immunological system, growth factors and metalloproteinases. In doing so, a number of researches seek for the investigation of genes differently expressed (up or down regulated) in endometriotic lesions cells using a variety of methodologies. The subtractive hybridization is a gene screening methodology that compares two distinct cellular groups, allowing the isolation of cDNA molecules representative from the genome of only one of the two groups, because it removes common sequences among them. The rapid subtractive hybridization methodology (RaSH) makes the process of subtraction easier and more efficient, identifying a large amount of differently expressed sequences. In this study, RaSH was used to identify differently expressed genes in 11 samples of endometriotic lesions (five from ovarian origin and six from peritoneum origin), and in 11 samples from topic endometrial tissue from women without the disease to determine possible molecular markers for diagnosis and treatment. After data analysis related to 166 reference sequences, genes HTRA1, LOXL1, SPARC and SSAT were selected for validation using real time quantitative RT-PCR. Only HTRA1 and LOXL1 presented significant difference between the two studied groups. Genes HTRA1 and LOXL1 were up regulated in endometriotic lesions from affected patients when compared with the expression levels in the endometrial tissue of non-affected women. Our results suggest that the genes HTRA1 and LOXL1 can be considered candidate molecular markers for the diagnosis of endometriosis.

Students studying organic chemistry often have difficulty applying prior knowledge from general chemistry in their thinking about organic reaction mechanisms. In the United States, electrophilic aromatic substitution (EAS) mechanisms can be taught towards the end in a second-semester course of organic chemistry, providing students with almost two-semesters' worth of experience with organic chemistry reactions before solving problems on synthesis of substituted aromatic compounds.<div>Little research has been done on how, or if, instructors consider their students' prior knowledge or understanding of these concepts in EAS in their teaching activities. The purpose of this study was to describe how students reason through EAS synthesis problems and to identify concepts or gaps in understanding that inhibit students from successfully solving these types of problems. Participants were interviewed using a think-aloud protocol in which they were asked to describe the reactants and mechanisms necessary to synthesize di- and tri-substituted benzenes using EAS. The interviews were transcribed and analyzed using a qualitative inquiry approach and the data interpreted in terms of the ACS Examinations Institute's Anchoring Concepts Content Maps for general and organic chemistry.</div><div>The findings from this study indicated that while students correctly applied their knowledge of substituent effects to solve these types of problems, they relied on rote-memorization of these effects, resulting in inflexibility when applying them to novel situations. Additionally, students exhibited gaps in understanding of fundamental concepts in resonance theory and Lewis structures, differentiating and utilizing Friedel-Crafts reactions, and recognizing when to use oxidation/reduction reactions in their syntheses.</div><div>Another component of this study focused on instructors of organic chemistry from a range of institutions in the United States. The purpose of this study was to describe how organic chemistry instructors perceived their students' reasoning about these types of problems, and to describe the characteristics of each instructors' topic-specific pedagogical content knowledge (TS-PCK) and the three general knowledge domains (GKDs) instructors draw upon to inform their TS-PCK. These knowledge domains are knowledge of students, subject matter knowledge, and pedagogical knowledge. These participants were remotely-interviewed using a think-aloud protocol in which they were asked to describe their classroom practices and teaching strategies when teaching EAS, and to describe how they would synthesize the same aromatic compounds as their students (a selection of which were interviewed in the previous study). Participants were asked to consider how their students would approach the syntheses and to specify what parts of the syntheses their students would find challenging, and why. The interviews were transcribed and analyzed using a qualitative inquiry approach. </div><div>The findings from this study indicated that the instructors were aware of their students' tendencies to use rote-memorization without understanding in the course, but there was still a mis-alignment between how instructors' perceived their students' reasoning through EAS synthesis problems and the reasoning the students actually used. The instructors believed that their students would only rely on the directing effects of substituents in their reasoning, but the students demonstrated they were aware of the activating and deactivating effects too. Additionally, instructors believed their students would not be hindered by an understanding of resonance or Lewis structures in their syntheses.</div><div>Finally, there are some recommendations for addressing the students' propensity for rote-memorization by providing a visual way to represent directing and activating/deactivating effects of substituents using electrostatic potential maps. There are also suggestions for further building on this work. <br><div><br></div></div>

The subject of dangerous goods as it pertains to carriage by sea is of growing importance and concern because it impacts on safety as well as environmental issues. Both involve liability associated with maritime transportation and liability in respect of dangerous goods is a complex area of law both from an international as well as a domestic perspective. China is a rapidly emerging economic power and a major world player in shipping and seaborne trade including import and export of hazardous substances. Furthermore, China is undergoing remarkable reform and transformation in all respects, and legal regimes, especially in the maritime field, are in a state of evolution. This thesis presents a two-fold area of concentration, that is, the international regime and the domestic Chinese law, looking at the safety as well as the environmental dimensions of international carriage of dangerous goods by sea. In order to carry out a comparative analysis of the international and Chinese legal regimes pertaining to the issues of contractual and tortious liability, a relatively detailed analytical examination of the international regime has been completed. Following this, the legal regime under Chinese law concerning the sea carriage of dangerous goods is critically evaluated in terms of the evolution of the domestic maritime law and the issues of application of international law and domestic law from the perspectives of regulatory law and civil liability. The discussion on the existing issues liability is centered on the principles of liability in tort and contract borne by private parties and state responsibility in respect of damage arising from the maritime transportation of dangerous goods. Conclusions are drawn from the summaries of chapters highlighting the critical issues in light of the findings of the research; the appropriate recommendations and suggestions for improvements to the international regimes; and proposals for law reform in the form of new legislation or amendments to existing legislation with the aim of improving the domestic regime to bring it into closer alignment with international law on the carriage of dangerous goods by sea.

O objetivo desta dissertação é descrever o comportamento de estruturas com repetição verbal em português brasileiro que apresentam características específicas diferentes de outras construções com repetição verbal já estudas no português brasileiro (Bastos 2001/ Bastos-Gee 2009) e no português europeu (Martins 2007). As estruturas em questão são iniciadas com que, sendo que em cada uma delas o constituinte encabeçado por que aparece em posições diferentes na sentença: quando o que está no primeiro bloco, nomeio a estrutura de que-reportativo; quando o que está no segundo bloco, nomeio a estrutura de que-adversativo. Essas construções apresentam similaridades, como a necessidade dos constituintes verbais serem idênticos, mas também apresentam muitas diferenças entre si além da posição do que: ter necessidade de repetir um antecedente discursivo obrigatoriamente, que só ocorre com que-reportativo, e a presença de uma sentença coordenada adversativa implícita, o efeito-mas (Bastos 2001/Bastos-Gee 2009), que também só ocorre com que-reportativo. Através da comparação com outras estruturas de repetição verbal já estudadas no português, mostro que as construções analisadas nesta dissertação possuem comportamentos diferenciados e envolvem tópicos diferentes, a saber, a estrutura com que-reportativo envolve tópico conversacional, enquanto a estrutura com que-adversativo envolve tópico contrastivo. / The goal of this dissertation is to describe the behavior of finite structures with verbal repetition in Brazilian Portuguese, which present different specific characteristics from other verbal repetition constructions previously studied in Brazilian Portuguese (Bastos 2001/ Bastos-Gee 2009) and in European Portuguese (Martins 2007). The structures in question start with que that, but the constituents headed by que occupy different positions in each structure: when que is in the first block of the sentence, I call the structure reportative-que; when que is in the second block of the sentence, I call the structure adversative-que. These structures present similarities, such as the requirement of identical verbal constituents, but also exhibit many differences besides the position where que appears: the requirement to repeat the previous discourse and the presence of an implicit adversative clause - the but-effect (Bastos 2001/ Bastos-Gee 2009) -, for instance, only happens with reportative-que. Through the comparison with other verbal repetition structures already studied in Portuguese, I demonstrate the structures analyzed in this dissertation have a differentiated behavior. I also demonstrate that they are composed by different topics, namely, reportative-que has a conversational topic, while adversative-que has a contrastive topic.

Conclusions. The retractable non-invasive needle seems to have reasonably good capability of blinding patients and is more successful in acupuncture-naive subjects than in experienced ones. The failure to produce similar sensation of stimulation in the sham acupuncture group may be the major reason for incomplete blinding. (Abstract shortened by UMI.) / Design, participants and outcomes. We conducted a randomized controlled trial of 398 acupuncture-naive subjects and 197 acupuncture-experienced subjects who were recruited from the outpatients clinic of a university hospital in China. Acupuncture-naive and experienced subjects were randomized separately to receive the real acupuncture treatment or the placebo acupuncture at acupoint Hegu on the left hand. After the intervention, the percentage of patients who thought they had received real acupuncture was compared between the real and sham acupuncture groups. We defined the degree of blinding as 100% minus the difference in the percentage between the two groups. The score of pain, soreness, numbness, heaviness and distension during the treatment was also compared between the two groups. / Introduction. The newly designed retractable needle which can be used to produce non-invasive placebo or (sham) acupuncture seems promising and has been used in clinical trials. Several studies investigated the credibility of the retractable needle in blinding patients. These studies are generally small and mostly in acupuncture-naive patients and mainly focused on the difference in acupuncture-related sensations rather than the successfulness of blinding. / Objectives. To quantify the degree of blinding of the retractable non-invasive placebo needle in blinding both acupuncture-naive and experienced subjects in clinical trials and to explore possible reasons for incomplete blinding of the sham needle. / Results. In acupuncture-naive subjects, the percentage of those who thought they had received real acupuncture was 70.4% and 42.7% respectively in the real and sham acupuncture groups. The degree of blinding was 72.3% (95% CI: 62.9%, 81.7%). In acupuncture-experienced subjects, the degree of blinding was 58.9 (95% CI: 46.2%, 71.6%). The difference in degree of blinding between acupuncture-naive and experienced subjects was statistically significant (P&lt;0.05). The score of pain, soreness, numbness, heaviness and distension in the real acupuncture group was all statistically significantly higher than that in the sham acupuncture group in both acupuncture-naive and experienced subjects (P&lt;0.001). / Zhang Hongwei. / "February 2005." / Adviser: Jin Ling Tang. / Source: Dissertation Abstracts International, Volume: 67-01, Section: B, page: 0161. / Thesis (Ph.D.)--Chinese University of Hong Kong, 2005. / Includes bibliographical references (p. 93-102). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstracts in English and Chinese. / School code: 1307.

Keywords: Non-inferiority Test, Bootstrap, Misclassification, Partially Validated Data. / Moreover, misclassification is frequently encountered in collecting ordinal categorical data. We also consider the non-inferiority test based on the data with misclassification. We have explored two different approaches. The first approach can be applied when misclassification probabilities are known or can be calibrated. The second approach deals with the case when we have partially validated data that provide the information on misclassification. The proposed approaches have wide applications that are not confined to tests in medical research. We design a substantive study to illustrate the practicality and applicability of the proposed approaches. / When a new treatment comes out, it is likely to find benefits of the new one, such as fewer side effects, greater convenience of employment, or lower cost in terms of money and time. Therefore, the more appropriate research question is whether the new one is non-inferior or equivalent to, but not necessarily superior to the reference treatment. Consequently, the non-inferiority test or equivalence test is widely used in medical research, which is oriented towards showing that the difference of effect between the two treatments probably lies in a tolerance interval with the pre-defined lower or upper bounds. In this thesis, we consider non-inferiority tests when the data are ordinal categorical. In particular, we are interested in correlated data. We will develop non-inferiority testing procedures for data that are obtained by the paired design and three-armed design. We take advantage of a latent normal distribution approach to model ordinal categorical data. / Han, Yuanyuan. / Adviser: Poon Wai-Yin. / Source: Dissertation Abstracts International, Volume: 73-06, Section: B, page: . / Thesis (Ph.D.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (leaves 114-117). / Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. [Ann Arbor, MI] : ProQuest Information and Learning, [201-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web. / Abstract also in Chinese.