Search

The multinomial model was fitted to the stated preference data of the transportation problem by commuters in Mamelodi, east of Pretoria. The data analyzed in the study was collected in 2001 among 151 less literate (highest level of education up to Standard 5 or Grade 7) and 194 literate (highest level of education from standard 6 or Grade 8 to Standard 10 or Grade 12) commuters in the CBD (Central Business District) of Pretoria. Seventeen (17) variables have been analyzed. The objective of the study is to determine if there are differences when three types of codings (dichotomous, binary and effect) are applied to the same data. The final interest is to determine those factors that affect commuters in choosing their mode of transport to work in the CBD of Pretoria. All the logistic regression models and multinomial logit models tested in the study were found to be statistically significant for the three different codings. Due the limitations that SAS has, the logistic regression models were fitted and used to carryout the analyses. When variables were selected by the stepwise procedure, and only those explanatory variables that were significant fitted in the model, the three models were all statistically significant for the Hosmer-Lemeshow goodness-of-fit statistics, but not for the Pearson and Deviance goodness-of-fit statistics., SANPAD (South African-Netherlands Programme on Alternatives in Development)

The goal of this paper is to improve the prediction performance of fault-prone module prediction models (fault-proneness models) by employing over/under sampling methods, which are preprocessing procedures for a fit dataset. The sampling methods are expected to improve prediction performance when the fit dataset is unbalanced, i.e. there exists a large difference between the number of fault-prone modules and not-fault-prone modules. So far, there has been no research reporting the effects of applying sampling methods to fault-proneness models. In this paper, we experimentally evaluated the effects of four sampling methods (random over sampling, synthetic minority over sampling, random under sampling and one-sided selection) applied to four fault-proneness models (linear discriminant analysis, logistic regression analysis, neural network and classification tree) by using two module sets of industry legacy software. All four sampling methods improved the prediction performance of the linear and logistic models, while neural network and classification tree models did not benefit from the sampling methods. The improvements of Fl-values in linear and logistic models were 0.078 at minimum, 0.224 at maximum and 0.121 at the mean., ESEM 2007 : First International Symposium on Empirical Software Engineering and Measurement, 20-21 Sept. 2007, Madrid, Spain

In this paper, we experimentally evaluated the effect of outlier detection methods to improve the prediction performance of fault-proneness models. Detected outliers were removed from a fit dataset before building a model. In the experiment, we compared three outlier detection methods (Mahalanobis outlier analysis (MOA), local outlier factor method (LOFM) and rule based modeling (RBM)) each applied to three well-known fault-proneness models (linear discriminant analysis (LDA), logistic regression analysis (LRA) and classification tree (CT)). As a result, MOA and RBM improved Fl-values of all models (0.04 at minimum, 0.17 at maximum and 0.10 at mean) while improvements by LOFM were relatively small (-0.01 at minimum, 0.04 at maximum and 0.01 at mean)., ESEM 2007 : First International Symposium on Empirical Software Engineering and Measurement, 20-21 Sept. 2007, Madrid, Spain

This version of the article has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/s41105-023-00470-4., As lifestyles have shifted to favor nighttime activities, daytime sleepiness and sleep-related problems have become increasingly common in Japan. Excessive daytime sleepiness (EDS) is an adverse consequence of sleep loss and an important public health concern. EDS may cause academic difficulties, behavioral abnormalities, and psychological dysfunction; therefore, it is a particularly important issue among university students. We conducted a cross-sectional study to investigate the prevalence of EDS and its associated lifestyle factors among Japanese university students. A questionnaire was completed by 1470 first-year university students, aged 19.0 (± 1.0) years. Using the questionnaire, we collected information on (1) demographic variables, (2) lifestyle variables, and (3) sleep habits and daytime sleepiness. Daytime sleepiness was measured using the Japanese version of the Epworth Sleepiness Scale, a frequently used subjective scale for assessing sleepiness. The overall prevalence of EDS was 57% (53% in men and 61% in women). Multivariate logistic regression analysis revealed that the following factors were associated with EDS: female sex, exercise habits, long commuting times, later wake-up times, and shorter sleep duration. Given that more than 50% of first-year university students reported having EDS, interventions should be considered to decrease its risk, including educational programs that provide strategies to extend sleep duration and delay wake-up time. Such strategies may also be valuable for students with other potential risk factors, such as exercise habits or long commute times, that are associated with EDS.

Bu kitap 2011 yılında İstanbul Üniversitesi Sosyal Bilimler Enstitüsü’nce kabul edilmiş Doktora tezimden türetilmiştir. Tez çalışmamı kitap olarak yayınlatma kararı almamdaki en önemli neden, çeşitli bilim dallarında araştırmacılar tarafından yürütülen çalışmalarda, bağımlı değişkenin nitel (kategorik) olduğu durumlarla karşılaşılması ve bu tür problemlerin analizinde, bağımlı ve bağımsız değişkenler arasındaki ilişkinin klasik regresyon modeli ile kurulmasının mümkün olmaması ve bu noktada problemlerin modellenebilmesi için alternatif yöntemlere ihtiyaç duyulmasıdır. Bu kitap ile amaçlanan, araştırmacılara bağımlı değişkenin kategorik olduğu problemlerle karşılaştıklarında, değişkenler arasındaki ilişkileri kurabilmek için kullanılabilecekleri birçok uygun istatistik yöntem arasından, akademik çalışmalarda genellikle en çok tercih edilen ve bu tür modellerin analizinde kullanılması uygun olan İki Kategorili Lojistik Regresyon Analizi ile İki Kategorili Diskriminant Analizi hakkında, faydalanabilecekleri ve çalışmalarına ışık tutabileceğine inandığım bilgileri sunmaktır. Bu kitabın ilk bölümü, iki alt başlıktan oluşmaktadır. İlk alt başlıkta, Genelleştirilmiş Doğrusal Modellerden ve Doğrusal Olasılık Modeli ile Lojit ve Probit modellerden kısaca bahsedilmiştir. İkinci alt başlıkta, Lojistik Regresyon Analizi için modelin varsayımları, model kurma stratejileri, parametre tahmin yöntemi, anlamlılık ve uyum iyiliği testleri ile analiz sonuçunda elde edilen model parametrelerinin nasıl yorumlanması gerektiğiyle ilgili teorik bilgilere yer verilmiştir. İkinci bölümde, Diskriminant Analizi, analizin varsayımları, diskriminant fonksiyonun elde edilmesi, elde edilen fonksiyonun test edilmesi ve yorumlanması ile ilgili genel bilgiler paylaşılmıştır. Ayrıca bu analizin ayırıcı gücünün ve dışsal geçerliliğin nasıl ele alınabileceğiyle ilgili teorik bilgiler de bu bölüm içinde yer almaktadır. Üçüncü bölümde, işletme performansını etkileyebilecek faktörlerin araştırılması başlığ

A field survey of indoor environmental measurements and questionnaires on thermal sensation, overall comfort, and behaviors was conducted in four office buildings in Japan by visiting each office every month over a duration of more than a year during the coronavirus disease 2019 (COVID-19) pandemic. The indoor environment was measured concurrently. We obtained 1047 votes from office workers in their 20s to 60s. The regression and Griffiths’ methods were used to calculate the indoor comfort temperature. A logistic regression analysis was used to develop the occupant behavior model. Over 70% of the occupants found the indoor environment comfortable at a mean comfort temperature of 23.2 to 25.9 °C. Gender differences were observed in thermal sensation and overall comfort, but a gender difference was observed only in the cooling mode for the indoor comfort temperature. An adaptive model was developed for the office buildings in Nagasaki city to predict the indoor comfort temperature from the outdoor air temperature. The proportions of heating, cooling, and fan usage can be predicted from the outdoor air temperature using a logistic regression analysis. The adaptive model and occupant behavior model are useful for the indoor temperature control of the existing buildings and thermal simulation of the new building design., Energies, 16(16), art. no. 6060; 2023

Background: Continuity of care constitutes the basis of primary health care services and is associated with decreased hospitalization. In Finland, accessibility to primary care and increased use of hospital services are recognized challenges for the health care system. Objectives: The aim of the study was to determine whether having a named GP is associated with hospital service use. Methods: The data are part of the Health and Social Support study (HeSSup) based on a random Finnish working-age population sample. The cohort of the study comprised participants of postal surveys in 1998 (n = 25,898) who returned follow-up questionnaires both in 2003 and 2012 (n = 11,924). Background characteristics were inquired in the questionnaires, and hospitalization was derived from national registries (Hilmo-register). Results: A named GP was reported both in 2003 and 2012 only by 34.3% of the participants. The association between hospital days and a named GP was linearly rising and statistically significant in a single predictor model. The strongest associations with hospital use were with health-related factors, and the association with a named GP was no longer significant in multinomial analysis. Conclusion: A named GP is associated with an increased use of hospital days, but in a multinomial analysis the association disappeared. Health related factors showed the strongest association with hospital days. From the perspective of the on-going Finnish health and social services reform, continuity of care should be emphasized., CC BY 4.0 DEEDCC0 1.0 DEED© The Author(s) 2023. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.Correspondence Address: E. Lautamatti; Faculty of Medicine and Health Technology, Tampere University, Tampere, Finland; email: emmi.lautamatti@tuni.fiOpen access funding provided by Tampere University (including Tampere University Hospital). A personal grant was awarded to the corresponding author by the Foundation of General Practice, General Practitioners in Finland (GPF), Tampere University Hospital, The Finnish Medical Association and The Finnish Medical Foundation (No. 7045).

In this thesis, we present the use of logistic regression method to develop a credit scoring modelusing the raw data of 4447 customers of a bank. The data of customers is collected under 14independent explanatory variables and 1 default indicator. The objective of this thesis is toidentify optimal coefficients. In order to clean data, the raw data set was put through variousdata calibration techniques such as Kurtosis, Skewness, Winsorization to eliminate outliers.On this winsorized dataset, LOGIT analysis is applied in two rounds with multiple statisticaltests. These tests aim to estimate the significance of each independent variable and modelfitness. The optimal coefficients can be used to obtain the credit scores for new customers witha new data set and rank them according to their credit risk.

The aim of this study is to contribute to the growing literature on e-participation by analysing various cases of participation. More detailed, the goal is to gain further insight into the role of perceived advantages with regards to citizen willingness in terms of digital and non-digital participation. While there are several studies identifying advantages of e-participation compared to traditional, non-digital alternatives, there is little research into how the perception of these advantages is influencing citizen motivation to engage in e-participation. This study examines three participation initiative cases with different levels of citizen engagement. We use a logistic regression model to analyse our data. The findings reveal new insights for both research and practice. First, the results show that simplicity, efficiency, location independence, and cost reduction are generally considered to be advantages of e-participation. In comparison, data security and data protection are advantages of non-digital participation. However, only cost reduction and simplicity have a positive influence on citizens’ willingness to engage in all three cases. Additionally, when data security was perceived as an advantage of e-participation, this also leads to a higher likelihood of preferring digital over non-digital participation. This is true for the two cases with higher engagement levels. These findings differ from previous studies and raise questions regarding the impact that the engagement level has on the results. By studying this topic further, valuable insights for governments on how to use and promote e-participation initiatives can be gained.

Introduction. One of the most significant indicators of quality in the academic endeavor of educational institutions is the academic performance of their student population. However, its study is complex, given its multi-causal nature. In view of the above, the general aim of this research was to identify those variables that could impact the academic performance (success or failure) of students admitted into a private university in Lima. Methodology. Two samples were taken corresponding to entrants of the two semesters of the academic year, for a total of 5040 participants (3515 from the first semester and 1889 from the second semester). The instruments applied were the Learning Strategies Questionnaire (Martínez Guerrero, 2004), the Vocational Self-Efficacy Sc ale (Carbonero Martín y Merino Tejedor, 2002), the Degree Maturity Questionnaire (Álvarez González et al., 2007), and the Test Anxiety Questionnaire (Valero Aguayo, 1999). Results. As part of the results of the logistic regression analysis, it was found that the variables that best predicted a failing weighted average were the following: a higher number of credits enrolled in leveling courses, male sex, lower score in self-efficacy beliefs, negative attitude towards the study program, a lower score in achievement motivation, and a higher score in exam anxiety. Discussion. It is concluded that academic performance can be explained from a set of motivational, emotional, and vocational variables, as well as some sociodemographic variables such as gender. Based on the results obtained, it is recommended to direct preventive actions toward those psychological and demographic aspects that are somehow associated with the performance of university entrants.

Statistical techniques have been improved over the years and can now be applied to practically anything, including sports. Sports analytics are a hot topic right now due to the huge amount of data available for analysis. This data can be analyzed for both improving the performance of the athletes and finding exogenous variables that may affect the results. This work aims to predict Women Tennis matches in the Grand Slams and understand what affects performances of the players and how. The objectives are interesting because they are chosen based on what a tennis fan would want to know. The statistical techniques used are descriptive to understand the performances, and predictive to predict the winners using six different models.

For remote communities, access to permafrost information for hazard assessment is a considerable challenge. This study applies analytical methods illustrating a time- and cost-efficient method for conducting community-scale permafrost mapping in the community of Whatì, NT. A binary logistic regression model was created using a combination of field data, digital elevation model-derived variables and remotely sensed products. Independent variables included categorical inputs such as vegetation, topographic position index and elevation breaks. The dependent variable is sourced from 139 physical checks of permafrost presence/absence. Vegetation was shown to be the strongest predictor of permafrost. The model predicts 50.0 % of the vegetated area is underlain by permafrost with a model accuracy of 91.4 % and spatial agreement of 72.8 % when compared to ground-truth pits. Compared to existing permafrost products this value is on the lowest edge of Whatì’s current classification (extensive discontinuous) illustrating there could be less permafrost than presumed.

Background: Hypertensive patients with primary aldosteronism (PA) have a higher risk of cardiovascular complications than those with blood pressure-matched essential hypertension. The excess cardiovascular consequences of PA can be attributed to the proinflammatory effect of excessive aldosterone and mineralocorticoid receptor activation in a range of peripheral tissues and cell types. The neutrophil-to-lymphocyte ratio (NLR) is a widely available marker of inflammation which has been shown to predict cardiovascular outcome in the general population. This study aims to evaluate the use of NLR as a potential biomarker of PA and PA severity. Method(s): Patients with PA (n = 355) were identified from 2 large PA databases in Australia and China, while controls (n = 222) were patients with hypertension who were referred for assessment but did not meet the diagnostic criteria for PA. The NLR was retrospectively collected from routine full blood examination, prior to commencement of targeted treatment for PA. Result(s): The NLR did not differ between PA patients and hypertensive controls (median 2.3 and 2.4, P = 0.563). However, among patients with PA, the NLR was positively correlated with baseline and post-saline aldosterone levels (r = 0.22 and P < 0.001 for both) and negatively correlated with serum potassium (r = -0.15, P = 0.006). Furthermore, in a logistic regression analysis of data from patients with PA, the NLR predicted the presence of comorbid chronic kidney disease (CKD) (defined as estimated glomerular filtration rate <60 mL/min/1.73m2) with an odds ratio of 1.5 (P = 0.003). Conclusion(s): While the NLR did not distinguish PA from controls, it was a marker of PA severity, being associated with aldosterone concentration as well as the presence of CKD. A prospective study is needed to further clarify the role of NLR in predicting end-organ damage associated with PA.Copyright © 2020 The Author(s).

Habitual tobacco and alcohol use among adolescents, as well as sedentary lifestyle, are a public health problem and a pediatricepidemic. Quality of life is an indicator of overall self-perceived health status. The objective was to analyze the relationship betweenphysical activity, tobacco consumption, alcohol and quality of life using an observational methodology, randomly selecting 1037 Spanishadolescents aged 14-17 years old (15.52 ± 1.11 years old). Instruments used were the International Physical Activity Questionnaire(physical activity), the Youth Risk Behavior Surveillance (tobacco), the Questionnaire on alcohol consumption habits in adolescents(alcohol), and the Health Questionnaire SF-36 (quality of life). Bi-variant relationships were explored using contingency tables, followedby binary logistic regression analysis. Results suggest that physical activity (p <.001) is the variable that best predicts quality of lifeperceived by boys, tobacco use being also relevant (p <.05). In girls, both tobacco consumption (p <.001) and alcohol consumption (p<.001) are shown as the variables that best predict their perceived quality of life, but the association between physical activity andquality of life is less relevant (p <.05). The promotion of health-related quality of life depends on the creation of comprehensiveeducational environments where physical activity is part of a healthy lifestyle., El consumo habitual de tabaco y alcohol entre los adolescentes, así como el sedentarismo constituyen un problema de saludpública y una epidemia pediátrica. La calidad de vida es un indicador del estado general de salud autopercibido. El objetivo fue analizarla relación entre actividad física, consumo de tabaco, alcohol y calidad de vida mediante una metodología observacional, seleccionando demanera aleatoria 1037 adolescentes españoles de 14-17 años de edad (15.52 ± 1.11 años). Los instrumentos empleados fueron elInternational Physical Activity Questionaire (actividad física), el Youth Risk Behavior Surveillance (tabaco), el Cuestionario sobrehábitos de consumo de alcohol en adolescentes (alcohol) y el Cuestionario de Salud SF-36 (calidad de vida). Se exploraron las relacionesbi-variantes mediante tablas de contingencia, seguido de un análisis de regresión logística binaria. Los resultados sugieren que la actividadfísica (p < .001) es la variable que mejor predice la calidad de vida percibida por los varones siendo también relevante el consumo de tabaco(p < .05). En las mujeres, tanto el consumo de tabaco (p < .001) como el consumo de alcohol (p < .001), se muestran como las variablesque mejor predicen la calidad de vida percibida, siendo en menor medida relevante la asociación entre actividad física y calidad de vida (p< .05). La promoción de la calidad de vida relacionada con la salud depende de la creación de ambientes educacionales integrales donde laactividad física forme parte de un estilo de vida saludable.

The present study aimed to analyze the physiological, physical, technical, and tactical responses of young soccer playersaccording to the result in small-sided games. The sample consisted of 39 players (15.9 ± 1.0 years of age) grouped by the outcome of thegames, winning (17), drawing (6), and losing (16). Physical and physiological responses were quantified using a portable GPS and acardiac monitor. The rating of perceived exertion was measured using the BORG 6-20 scale. Technical-tactical responses were assessedaccording to the frequency and quality of execution of technical skills and core tactical principles. Shapiro-Wilk, Kruskal-Wallis, MannWhitney U and Multinomial Logistic Regression tests were used in the analysis. Significance was set at 5%. Players in the groupswinning and drawing covered greater distances in low-intensity running and shorter distances in walking. The drawing group executedmore passes than the losing group (14.0 vs 10.0). The winning group showed greater efficiency of execution in three offensive and threedefensive principles compared to the losing group. It was found that the increase in the efficiency of execution of offensive and defensiveprinciples increases the chances of the players being in drawing (11 to 24%) or winning (10% to 21%) groups in relation to losing. It isconcluded that the physical, technical, and tactical responses influence competitive success, and the increase in tactical efficiency resultsin more chances to achieve winning or drawing, O presente estudo objetivou analisar as respostas fisiológicas, físicas, técnicas e táticas de jovens futebolistas em função doresultado em jogos reduzidos. A amostra foi constituída por 39 jogadores (idade 15,9 ± 1,0 anos) agrupados pelo desfecho dos jogos,vitória (17), empate (6) e derrota (16). As respostas físicas e fisiológicas foram quantificadas por meio da utilização de um GPS portátile um monitor cardíaco. A percepção subjetiva de esforço foi medida utilizando a escala de BORG 6-20. As respostas técnico-táticas foramavaliadas conforme a frequência e qualidade de execução de habilidades técnicas e princípios táticos fundamentais. Na análise foramempregados os testes de Shapiro-Wilk, Kruskal-Wallis, U de Mann Whitney e a Regressão Logística Multinomial. A significância foiestabelecida em 5%. Jogadores que conquistaram a vitória ou o empate percorreram maiores distâncias em corrida de baixa-intensidadee menores distâncias em caminhada. O grupo empate realizou mais passes do que o grupo derrota (14,0 vs 10,0). O grupo vencedorapresentou maior eficiência de execução em três princípios ofensivos e três princípios defensivos em comparação com o grupo perdedor.Verificou-se que o aumento na eficiência de execução de princípios ofensivos e defensivos aumenta as chances de os jogadores estarem nosgrupos empate (11 a 24%) ou vitória (10% a 21%) com relação a derrota. Conclui-se que as características físicas, técnicas e táticas sãofatores que exercem influência sobre o êxito competitivo, sendo que o aumento na eficiência tática resulta em mais chances para conquistada vitória ou empate., El presente estudio tuvo como objetivo analizar las respuestas fisiológicas, físicas, técnicas y tácticas de los futbolistas jóvenes dependiendo del resultado en juegos reducidos. La muestra consistió en 39 jugadores (edad 15.9 ± 1.0 años) agrupados por el resultado de los juegos, victoria (17), empate (6) y derrota (16). Las respuestas físicas y fisiológicas se cuantificaron utilizando un GPS portátil y un monitor cardíaco. La percepción subjetiva del esfuerzo se midió utilizando la escala BORG 6-20. Las respuestas técnicotácticas se evaluaron de acuerdo con la frecuencia y calidad de ejecución de las habilidades técnicas y los principios tácticos fundamentales. Shapiro-Wilk, Kruskal-Wallis, Mann Whitney U y pruebas de regresión logística multinomial se utilizaron en el análisis. La significancia se estableció en 5%. Los jugadores que han ganado la victoria o el empate han cubierto distancias más grandes en carreras de baja intensidad y distancias más cortas al caminar. El grupo de empate hizo más pases que el grupo de derrota (14.0 vs 10.0). El grupo victoria mostró una mayor eficiencia de ejecución en tres principios ofensivos y tres defensivos en comparación con el grupo derrota. Se encontró que el aumento en la eficiencia de la ejecución de los principios ofensivos y defensivos aumenta las posibilidades de que los jugadores estén en el grupo empate (11 a 24%) o victoria (10% a 21%) en relación con la derrota. Se concluye que las respuestas físicas, técnicas y tácticas influye en el éxito competitivo, y el aumento de la eficiencia táctica da como resultado más posibilidades de lograr la victoria o el empate.

Purpose: Dissatisfaction and an impaired quality of life after total knee arthroplasty (TKA) is often associated with postsurgical anterior knee pain (AKP). The underlying pathological mechanisms are not yet fully understood. Therefore, a multifactorial approach encompassing clinical and radiological parameters seemed reasonable and promising to investigate postsurgical AKP. Methods: In this cross-sectional study, 25 patients without and 25 patients with postsurgical AKP after unilateral TKA were randomly recruited from a larger cohort of patients. Multiple clinical and radiological parameters—including real-time shear wave elastography (SWE) to measure the patellar and quadriceps tendon elasticity—were acquired and subsequently associated with AKP. For statistical analysis, SPSS (IBM, version 25) was used. Results: In total 50 participants (58.0% men, mean age 63.42 years, mean body mass index 29.75 kg/m2), having different prosthetic designs implanted, were included. Independently of key covariates, the strength of the quadriceps muscle (p = 0.021), a thinner inlay (p = 0.041), and a lower position of the patella (p = 0.041) were associated with AKP. Although no correlation with AKP was found (p = 0.346, resp. p = 0.154), we observed significantly decreased Young’s modulus of the patellar and quadriceps tendons for the involved knee compared to the uninvolved knee (p < 0.001). Conclusion: In conclusion, quadriceps muscle strength, inlay thickness, and the patella position might be of particular relevance in avoiding postsurgical AKP. Future studies with larger sample sizes are needed to clarify the impact of quadriceps muscle strength and the postoperative patella position as well as the role of SWE as a personalized modifiable prediction marker.

Purpose: To identify predictive factors for postoperative uncorrected visual acuity (UCVA) following diffractive multifocal intraocular lens implantation. Subject and Method: Forty five eyes of 28 patients who underwent cataract surgery with diffractive multifocal intraocular lens TECNISR ZMB00 (AMO) implantation from January 2015 to March 2018 were included. We formed two groups randomly with 18 eyes of 11 patients defined as the training set for preoperative factors and 27 eyes of 17 patients defined as the validation set for postoperative effects. All eyes had undergone superior corneal incisions of 2.4 mm. The predictive factors included age, mean corneal refractive power, corneal astigmatism, spherical equivalent, coma and spherical aberration of cornea, depth of angle recess, mesopic and photopic pupil diameters, PDist and MDist (distances from the alignment light to the photopic and the mesopic pupil centers, respectively). The good-UCVA group was defined as patients who had logMAR ≦ 0 one month after the surgery. Independent-related factors were identified by stepwise logistic regression analyses. Results: Only PDist was adopted (P = 0.0108) in the training set. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was 0.91. In the validation set, it had a sensitivity of 66%, a specificity of 100% and a cut-off value of 0.23. Conclusions: It is supposed that distances from the alignment light to the PDist influence the postoperative UCVA following multifocal diffractive intraocular lens implantation., identifier:Acta medica Nagasakiensia, 62(3), pp.101-106; 2019

The survey investigated through the survey of federated athletes of Brazilian artistic gymnastics who participated in at least one of the two national championships (n = 620), whether the population size (stratified into five categories) is a factor that influence the sporting success, using binary logistic regression. Second, using the Kruskall-Wallis test, ranking the ranking of athletes in quartiles sought to identify whether population density is an influencing factor for sports performance. The results indicate that in Brazil, belonging to larger cities (100001 to 500000 inhabitants) makes athletes have 7.51 times (95% CI = 6.30-8.96) more chances to have sportive performance results, especially men in cities. with a lower population density compared to women, and it is possible to explore the influence of sport-promoting public policies in these cities., A pesquisa investigou por intermédio do levantamento dos atletas federados da ginástica artística brasileira que participaram de ao menos um dos dois campeonatos nacionais (n=620), se o tamanho da população (estratificada em cinco categorias) é um fator que influenciam no sucesso esportivo, utilizando para tanto regressão logística binaria. Por segundo, utilizando o teste de Kruskall-Wallis, classificando o ranking de atletas em quartis buscou identificar se a densidade populacional é um fator de influência para ter rendimento esportivo. Os resultados indicam que no Brasil, pertencer a cidades maiores (100001 até 500000 habitantes) faz com que atletas possuam 7,51 vezes (IC95%=6.30-8.96) mais chances de ter resultados esportivos de rendimento, com destaque para os homens em cidades com densidade populacional mais baixa em comparação com as mulheres, sendo um possibilidade explorar a influência das políticas públicas de icentivo ao esporte nestas cidades.

Measurement error and misclassification in covariates is always inevitable in data collection process. Treating variables measured with error as true value to build up regression models and perform statistical analysis can lead to misleading results. To get more accurate estimation results, the effect of measurement error and misclassification has to be considered. In this thesis, we mention some correction methods to estimate the parameters in linear and logistic regressions on error-prone covariates and one two-level categorical variable with misclassification. Simulation studies are made to compare the performance of these estimators for a finite sample. Finally, we present a real-data example on logistic regression.

Exposure to natural outdoor environments (NOE) has been shown in population-level studies to reduce anxiety and psychological distress. This study investigated how exposure to one's everyday natural outdoor environments over one week influenced mood among residents of four European cities including Barcelona (Spain), Stoke-on-Trent (United Kingdom), Doetinchem (The Netherlands) and Kaunas (Lithuania). Participants (n = 368) wore a smartphone equipped with software applications to track location and mood (using mobile ecological momentary assessment (EMA) software), for seven consecutive days. We estimated random-effects ordered logistic regression models to examine the association between mood (positive and negative affect), and exposure to green space, represented by two binary variables indicating exposure versus no exposure to NOE using GPS tracking and satellite and aerial imagery, 10 and 30 min prior to participants’ completing the EMA. Models were adjusted for home city, day of the week, hour of the day, EMA survey type, residential NOE exposure, and sex, age, education level, mental health status and neighbourhood socioeconomic status. In addition, we tested for heterogeneity of effect by city, sex, age, residential NOE exposure and mental health status. Within 10 min of NOE exposure, compared to non-exposure, we found that overall there was a positive relationship with positive affect (OR: 1.39, 95% CI: 1.06, 1.81) of EMA surveys, and non-significant negative association with negative affect (OR: 0.80, 95% CI: 0.58, 1.10). When stratifying, associations were consistently found for Stoke-on-Trent inhabitants and men, while findings by age group were inconsistent. Weaker and less consistent associations were found for exposure 30 min prior to EMA. Our findings support increasing evidence of psychological and mental health benefits of exposure to natural outdoor environments, especially among urban populations such as those included in our study.

In Statistics, most theories are developed to deal with statistical modeling and analysis when observations are assumed to have true values of covariates. However, for the EPIC-InterAct Study, which is a large prospective case-cohort study nested within the European Prospective Investigation into Cancer and Nutrition (EPIC), covariate measured with error is an inevitable problem since dietary intake assessed with a food frequency questionnaire is prone to measurement error. Ignoring measurement error in covariates will cause biases of estimates as well as loss of power, and sometimes leads to misleading results of data analysis. Thus, to obtain more reliable results, data analysis of incorporating measurement error model in the EPIC-InterAct Study is necessary. In this thesis, we consider an additive error model according to the characteristics of the EPIC-InterAct Study data, that is, 24HR intake measurements are biased and errors in FFQ and 24HR measurements are highly correlated. Using method of moments, we show how to estimate all unknown parameters of the error model as well as other useful statistics under three reasonable assumptions. By calculating skewness and kurtosis, we find out best transformation for interested nutrients in the EPIC-InterAct Study, to make them close to normal distribution. Based on our proposed additive error model, an approximate maximum likelihood estimation (AMLE) for covariate with measurement error under the logistic regression model is developed. This method can be regarded as the adjustment of the regression calibration method and can provide approximate consistent estimator. Asymptotic normality of this estimator is established under regularity conditions. Furthermore, we explore the estimator for the interaction term between one continuous variable and one categorical variable. Similarly, under the Cox regression model, we propose an approximate profile likelihood estimation (APLE) to deal with covariate measurement error in

Aims: Inverse associations between vitamin D status and risk of type 2 diabetes observed in epidemiological studies could be biased by confounding and reverse causality. We investigated the prospective association between vitamin D status and type 2 diabetes and the possible role of reverse causality. Method(s): We conducted a case-cohort study within the Melbourne Collaborative Cohort Study (MCCS), including a random sample of 628 participants who developed diabetes and a sex-stratified random sample of the cohort (n = 1884). Concentration of 25-hydroxyvitamin D (25(OH)D) was measured using liquid chromatography-tandem mass spectrometry in samples collected at recruitment. Logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) for the risk of type 2 diabetes for quartiles of 25(OH)D relative to the lowest quartile and per 25 nmol/L increase in 25(OH)D, adjusting for confounding variables. Result(s): The ORs for the highest versus lowest 25(OH)D quartile and per 25 nmol/L increase in 25(OH)D were 0.60 (95% CI: 0.44, 0.81) and 0.76 (95% CI: 0.63, 0.92; p = 0.004), respectively. In participants who reported being in good/very good/excellent health approximately four years after recruitment, ORs for the highest versus lowest 25(OH)D quartile and per 25 nmol/L increase in 25(OH)D were 0.46 (95% CI: 0.29, 0.72) and 0.71 (95% CI: 0.56, 0.89; p = 0.003), respectively. Conclusion(s): In this sample of middle-aged Australians, vitamin D status was inversely associated with the risk of type 2 diabetes, and this association did not appear to be explained by reverse causality.Copyright © 2018 Elsevier B.V.

Background: Currently, little evidence exists to guide mode of delivery in pregnant women with super obesity. There is a trend toward elective caesarean delivery in this population with little evidence of improved maternal or neonatal outcomes. Aim(s): Our study compares maternal and neonatal outcomes based on planned mode of delivery and aimed to identify predictors for a successful vaginal delivery. Material(s) and Method(s): This retrospective observational study explored maternal and neonatal outcomes of women with a body mass index >=50 who birthed following a singleton pregnancy at >=36 weeks gestation over a 10 year period at a single centre in Melbourne, Australia. Outcomes between women having a planned vaginal or planned caesarean birth were compared on an intention-to-treat basis, with logistic regression used to determine factors predictive of a vaginal birth. Result(s): A total of 275 women with super obesity were identified. One hundred and ninety-nine (72%) planned a vaginal birth, which was successful for 70%. Planned vaginal birth was associated with lower rates of postpartum complications requiring readmission (5% vs 16%; P = 0.003). Neonatal resuscitation requirements and Apgar scores at five minutes were similar between groups. Multiparity was the strongest predictor of a successful vaginal birth while medical intervention in labour was associated with a reduced rate of success in primiparous women. Conclusion(s): Contrary to current trends in practice, a trial of labour for women with super obesity, when facilities and skills permit, appears a safe option for mother and baby.Copyright © 2018 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists

Background: Severe rhabdomyolysis is associated with acute kidney injury, but it is unclear if patients developing rhabdomyolysis after illicit drug use have a higher risk of acute kidney injury compared to other causes. Aim(s): To provide a descriptive analysis of patients admitted with rhabdomyolysis, with a focus on illicit drug use, and to determine if illicit drug use was an independent predictor for acute kidney injury or renal replacement therapy. Method(s): We conducted a 5-year cohort study of patients admitted to Monash Health, a tertiary referral hospital network. We identified adult patients with muscle injury from ICD-10 AM codes, serum creatine kinase level greater than 1000 U/mL, and a clinical history consistent with rhabdomyolysis. We determined the prevalence and type of illicit drug involved and determined the association between illicit drug use and renal outcomes by logistic regression. Result(s): Of 643 patients, illicit drug use was identified in 12%. Acute kidney injury developed in 51%, and 5% required renal replacement therapy. Compared to the rest of the cohort, patients who used illicit drugs were younger and had higher peak serum creatine kinase, and developed a higher severity of acute kidney injury. In multivariable analysis, the factors associated with acute kidney injury were illicit drug use, peak creatine kinase, cardiovascular disease, concurrent sepsis and a clinically-evident pressure injury. Chronic kidney disease and need for fasciotomy were additional risk factors for renal replacement therapy. Conclusion(s): Illicit drug use was associated with acute kidney injury and renal replacement therapy independent of creatine kinase levels.Copyright © 2019 Royal Australasian College of Physicians

Background: Little is known about the association between socioeconomic status and long-term stroke outcomes, particularly in low- and middle-income countries. Method(s): Patients were recruited from the Mashhad Stroke Incidence Study in Iran. We identified different socioeconomic variables including the level of education, occupation, household size, and family income. Residential location according to patient's neighbourhood was classified into less privileged area (LPA), middle privileged area and high privileged area (HPA). Using Cox regression, competing risk analysis and logistic regression models, we determined the association between socioeconomic status and 1- and 5-year stroke outcomes. Generalized linear model was used for adjusting associated variables for stroke severity. Result(s): Six hundred twenty-four patients with first-ever stroke were recruited in this study. Unemployment prior to stroke was associated with an increased risk of 1- and 5-year post-stroke mortality (1 year: adjusted hazard ratio [aHR] 3.3; 95% CI 1.6-7.06: p = 0.001; 5 years: aHR 2.1; 95% CI 1.2-3.6: p = 0.007). The 5-year mortality rate was higher in less educated patients (<12 years) as compared to those with at least 12 years of schooling (aHR 1.84; 95% CI 1.05-3.23: p = 0.03). Patients living in LPA compared to those living in HPAs experienced a more severe stroke at admission (aB 3.84; 95% CI 0.97-6.71, p = 0.009) and disabling stroke at 1 year follow-up (OR 6.1; 95% CI 1.3-28.4; p = 0.02). Conclusion(s): A comprehensive stroke strategy should also address socioeconomic disadvantages.Copyright © 2019 S. Karger AG, Basel. All rights reserved.

Aims/hypothesis: The aim of this work was to investigate clinical outcomes following severe hypoglycaemia requiring prehospital emergency medical services (EMS) management. Method(s): We carried out a prospective, observational study of adults with diabetes attended by prehospital EMS for management of severe hypoglycaemia between April 2016 and July 2017. Information on precipitants, hospitalisation, length of hospital stay and recurrence was collected at 1 and 3 months following the episode of severe hypoglycaemia. Median and logistic regression models examined predictive factors. Result(s): Five hundred and five adults (61% male, median age 67 years) participated in the study. Fifty-two per cent had type 1 diabetes, 43% type 2 diabetes and 5% were unsure of their diabetes type. Following EMS management of the index episode of severe hypoglycaemia, 50.3% were transported to hospital. Of those transported, 41.3% were admitted to hospital for ongoing management (20.8% of all participants). The following factors predicted hospital admission: older age (OR 1.28 [95% CI 1.02, 1.60] per 10 years), greater number of comorbidities (OR 1.27 [95% CI 1.08, 1.48] per morbidity), moderate-severe injury accompanying the hypoglycaemia (OR 5.24 [95% CI 1.07, 25.8] compared with nil-mild injury) and unknown cause of hypoglycaemia (OR 2.21 [95% CI 1.24, 3.94] compared with known cause). The median (interquartile range) length of hospital stay was 4 (2-7) days. During follow-up, recurrent severe hypoglycaemia attended by prehospital EMS was experienced by 10.7% of participants. Predictive factors of recurrent severe hypoglycaemia in 3 months were decreased HbA1c (OR 1.97 [95% CI 1.27, 3.06] per 10 mmol/mol decrease) and a greater number of antecedent severe hypoglycaemia episodes (OR 1.12 [95% CI 1.03, 1.23] per episode). Conclusions/interpretation: Following an episode of severe hypoglycaemia managed by EMS, one-fifth of participants required hospital admission, more likely in thos

Background: Timely revascularization with percutaneous coronary intervention (PCI) reduces death following myocardial infarction. We evaluated if a sex gap in symptom-to-door (STD), door-to-balloon (DTB), and door-to-PCI time persists in contemporary patients, and its impact on mortality. Methods and Results: From 2013 to 2016 the Victorian Cardiac Outcomes Registry prospectively recruited 13 451 patients (22.5% female) from 30 centers with ST-segment-elevation myocardial infarction (STEMI, 47.8%) or non-ST-segment-elevation myocardial infarction (NSTEMI) (52.2%) who underwent PCI. Adjusted log-transformed STD and DTB time in the STEMI cohort and STD and door-to-PCI time in the NSTEMI cohort were analyzed using linear regression. Logistic regression was used to determine independent predictors of 30-day mortality. In STEMI patients, women had longer log-STD time (adjusted geometric mean ratio 1.20, 95% CI 1.12-1.28, P<0.001), log-DTB time (adjusted geometric mean ratio 1.12, 95% CI 1.05-1.20, P=0.001), and 30-day mortality (9.3% versus 6.5%, P=0.005) than men. Womens' adjusted geometric mean STD and DTB times were 28.8 and 7.7 minutes longer, respectively, than were mens' times. Women with NSTEMI had no difference in adjusted STD, door-to-PCI time, or early (<24 hours) versus late revascularization, compared with men. Female sex independently predicted a higher 30-day mortality (odds ratio 1.67, 95% CI 1.11-2.49, P=0.01) in STEMI but not in NSTEMI. Conclusion(s): Women with STEMI have significant delays in presentation and revascularization with a higher 30-day mortality compared with men. The delay in STD time was 4-fold the delay in DTB time. Women with NSTEMI had no delay in presentation or revascularization, with mortality comparable to men. Public awareness campaigns are needed to address women's recognition and early action for STEMI.Copyright © 2019 The Authors. Published on behalf of the American Heart Association, Inc., by Wiley.

Type 2 diabetes mellitus (T2DM) is a common complex trait arising from interactions among multiple environmental, genomic, and postgenomic factors. We report here the first attempt to investigate the association between immunoglobulin G (IgG) N-glycan patterns, T2DM, and their clinical risk factors in an Australian population. N-glycosylation of proteins is one of the most frequently observed co- and post-translational modifications, reflecting, importantly, the real-time status of the interplay between the genomic and postgenomic factors. In a community-based case-control study, 849 participants (217 cases and 632 controls) were recruited from an urban community in Busselton, Western Australia. We applied the ultraperformance liquid chromatography method to analyze the composition of IgG N-glycans. We then conducted Spearman's correlation analyses to explore the association between glycan biomarker candidates and clinical risk factors. We performed area under the curve (AUC) analysis of the receiver operating characteristic curves by fivefold cross-validation for clinical risk factors, IgG glycans, and their combination. Two directly measured and four derived glycan peaks were significantly associated with T2DM, after correction for extensive clinical confounders and false discovery rate, thus suggesting that IgG N-glycan traits are highly correlated with T2DM clinical risk factors. Moreover, adding the IgG glycan profiles to fasting blood glucose in the logistic regression model increased the AUC from 0.799 to 0.859. The AUC for IgG glycans alone was 0.623 with a 95% confidence interval 0.580-0.666. In addition, our study provided new evidence of diversity in T2DM complex trait by IgG N-glycan stratification. Six IgG glycan traits were firmly associated with T2DM, which reflects an increased proinflammatory and biological aging status. In summary, our study reports novel associations between the IgG N-glycome and T2DM in an Australian population and the putative r

Background: Currently, little evidence exists to guide mode of delivery in pregnant women with super obesity. There is a trend toward elective caesarean delivery in this population with little evidence of improved maternal or neonatal outcomes. Aim(s): Our study compares maternal and neonatal outcomes based on planned mode of delivery and aimed to identify predictors for a successful vaginal delivery. Material(s) and Method(s): This retrospective observational study explored maternal and neonatal outcomes of women with a body mass index >=50 who birthed following a singleton pregnancy at >=36 weeks gestation over a 10 year period at a single centre in Melbourne, Australia. Outcomes between women having a planned vaginal or planned caesarean birth were compared on an intention-to-treat basis, with logistic regression used to determine factors predictive of a vaginal birth. Result(s): A total of 275 women with super obesity were identified. One hundred and ninety-nine (72%) planned a vaginal birth, which was successful for 70%. Planned vaginal birth was associated with lower rates of postpartum complications requiring readmission (5% vs 16%; P = 0.003). Neonatal resuscitation requirements and Apgar scores at five minutes were similar between groups. Multiparity was the strongest predictor of a successful vaginal birth while medical intervention in labour was associated with a reduced rate of success in primiparous women. Conclusion(s): Contrary to current trends in practice, a trial of labour for women with super obesity, when facilities and skills permit, appears a safe option for mother and baby.Copyright © 2018 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists

This article describes how a data mining model was obtained and applied to the problem of university dropout in the Systems Engineering program of the University of Cundinamarca, in Facatativá. The model was structured by means of the KDD (knowledge discovery in databases) data mining methodology using Python programming language, Pandas data processing library, and the Sklearn machine learning. For the process, we took into account problems that are additional to the ones specific to the mining process, such as high dimensionality, reason why the methods of selection of the univariate statistical variables, feature importance, and SelectFromModel (Sklearn) were applied. In the project, five data mining techniques were selected for evaluation: nearest neighbors (KNN), decision tree (DT), random forest (RF), logistic regression (LR), and support vector machines (SVM). Regarding the selection of the final model, the results of each model were tested on the precision metrics, confusion matrix, and additional metrics of the confusion matrix. Finally, the parameters of the selected model were adjusted and the generalization of the model was evaluated by plotting its learning curve., Cet article décrit l'obtention d'un modèle d’exploitation des données appliqué à la problématique du décrochage universitaire du cursus d'ingénierie informatique du campus de Facatativá de l'université de Cundinamarca. Le modèle a été structuré à l'aide de la méthodologie d'exploitation de données KDD (knowledge discovery in databases) utilisant le langage de programmation Python, la bibliothèque de traitement de données Pandas et l'apprentissage automatique Sklearn. Lors de ce processus, des indicateurs supplémentaires d'extraction ont été retenus comme la dimensionnalité élevée, les méthodes de sélection des variables statistiques univariées ou l'importance des caractéristiques du SelectFromModel (Sklearn). Dans ce projet, cinq techniques d'exploitation des données ont été sélectionnées et soumises à évaluation: les parentés les plus proches (KNN), les matrices de décision (DT), les arbres aléatoires (RF), la régression logistique (LR) et les supports de machines vectorielles (SVM). Pour la sélection du modèle final, chaque résultat des différents modèles a été évalué grâce aux métriques de précision, à la matrice de confusion et aux métriques supplémentaires de la matrice de confusion. Enfin, les paramètres du modèle sélection of the final model, the results of each model were tested on the precision metrics, confusion matrix, and additional metrics of the confusion matrix. Finally, the parameters of the selected model were adjusted and the generalization of the model was evaluated by plotting its learning curve., Este artigo descreve a obtenção de um modelo de mineração de dados aplicado ao problema da evasão universitária no curso de Engenharia de Sistemas da Universidade de Cundinamarca, campus Facatativá. O modelo foi estruturado utilizando a metodologia de mineração de dados KDD (Knowledge Discovery in Databases) utilizando também a linguagem de programação Python, a biblioteca de processamento de dados Pandas e a Machine Learning Sklearn. Para o processo, foram levados em consideração os problemas adicionais ao processo de mineração, como, por exemplo, a alta dimensão, para a qual foram aplicados os métodos de seleção das variáveis estatísticas univariadas, feature importance e SelectFromModel (Sklearn). No projeto, foram selecionadas cinco técnicas de mineração de dados para avaliá-las: vizinhos mais próximos (K Nearest Neighbors, KNN), árvores de decisão (DT), árvores aleatórias (Random Forest, RF), regressão logística, (Logistic Regression LR) e máquinas de vetores de suporte (Support Vector Machines, SVM). No que diz respeito à seleção do modelo final, foram avaliados os resultados de cada modelo nas métricas de precisão, matriz de confusão e métricas adicionais da matriz de confusão. Por fim, os parâmetros do modelo selecionado foram ajustados e a generalização do modelo foi avaliada no momento de diagramar sua curva de aprendizado., En el presente artículo se describe la obtención de un modelo de minería de datos aplicado al problema de la deserción universitaria en el programa de Ingeniería de Sistemas de la Universidad de Cundinamarca, extensión Facatativá. El modelo se estructuró mediante la metodología de minería de datos KDD (knowledge discovery in databases) haciendo uso del lenguaje de programación Python, la librería de procesamiento de datos Pandas y de machine learning Sklearn. Para el proceso se tuvieron en cuenta problemas adicionales al proceso de minería, como, por ejemplo, la alta dimensionalidad, por lo cual se aplicaron los métodos de selección de las variables estadístico univariado, feature importance y SelectFromModel (Sklearn). En el proyecto se seleccionaron cinco técnicas de minería de datos para evaluarlas: vecinos más cercanos (K nearest neighbors, KNN), árboles de decisión (decision tree, DT), árboles aleatorios (random forest, RF), regresión logística (logistic regression, LR) y máquinas de vectores soporte (support vector machines, SVM). Respecto a la selección del modelo final se evaluaron los resultados de cada modelo en las métricas de precisión, matriz de confusión y métricas adicionales de la matriz de confusión. Por último, se ajustaron los parámetros del modelo seleccionado y se evaluó la generalización del modelo al graficar su curva de aprendizaje.

Nowadays, the development of technology has led to a large amount of ever-expanding and ever-updating datastreams. Therefore, it is necessary to develop statistical methodologies which can adjust the model timely and reflect the online manner of the ongoing process. To meet this issue, we construct an online classification system which can provide the classification of the data in time. This classification system consists of two steps. Firstly, a time-varying coefficient model is implemented to develop the estimation procedure, which can self-adjust its parameters over time and describe the dynamic feature of the data. Then the cost-sensitivity analysis method is introduced to the classification part. Unlike the traditional classification method, the classification method in our study emphasizes the difference among misclassification penalties, which satisfies the demand of the real-world application adequately. Under some mild conditions, the consistency of these estimators is established. Both the finite sample simulation and the real data application show that this methodology performs well. Key words: Varying Coefficient Model , Multinonomial Logistic Regression, Cost-Sensitivity Analysis.

Objective To develop a combinatorial panel of salivary cytokines that manifests the presence of non-small cell lung cancer (NSCLC) that will eventually improve prognosis by facilitating the early diagnosis and management of this common cancer. Methods We performed a case-control study comparing salivary cytokine profiles of 35 adult subjects with NSCLC with those of 35 matched, healthy nonsmokers. Multiplex bead array assays were used to quantify 27 cytokines in saliva, serum, and oral mucosal transudate samples. Logistic regression analysis was used to develop an informative cytokine panel. Receiver operating characteristic (ROC) curves were generated to evaluate the discriminant ability of the panel. Results A combinatorial 12-cytokine panel (interleukin receptor antagonist [IL1RN], IL1B, IL6, IL7, IL8, IL10, C-C motif chemokine ligand 11 [CCL11], tumor necrosis factor, C-X-C motif chemokine ligand 10 [CXCL10], C-C motif chemokine ligand 3, C-C motif chemokine ligand 4, and platelet-derived growth factor-BB) distinguished patients with NSCLC from healthy controls. Further, ROC analysis revealed that a cytokine panel comprising IL10 (odds ratio, 1.156) and CXCL10 (odds ratio, 1.000) discriminated NSCLC with a sensitivity of 60.6% and specificity of 80.8% (area under the ROC curve, 0.701). Conclusion A combinatorial panel of select salivary cytokines indicates the presence of NSCLC.

Objectives: Demoralization as a form of existential distress involves poor coping, low morale, hopelessness, helplessness, and meaninglessness. In a secondary analysis of a cohort of German cancer patients, we aimed to explore latent class structure to assess the contribution that symptoms of demoralization make to anhedonic depression, anxiety, adjustment, and somatic disorders. Method(s): Measures of demoralization, depression, anxiety, physical symptoms, and functional impairment had been completed cross-sectionally by 1527 patients with early or advanced cancer. Latent class analysis used maximum likelihood techniques to define the unobserved latent constructs that can be predicted as symptom clusters. Individual patients were assigned to the most probable class. Classes were compared on demographics, and logistic regression assessed the odds of individual items predicting each class. Result(s): A 4-class model provided the best fit. Class 1 (n = 829, 54.3%) was defined by the absence of distress; Classes 2 to 4 all carried functional impairment. Class 2 (n = 333, 21.8%) was differentiated by somatic symptoms (sleep, tiredness, and appetite); Class 3 (n = 163, 10.7%) by anhedonia, anxiety, and severe demoralization; and Class 4 (n = 202, 13.2%) by adjustment and moderate demoralization. Members of Class 3 were more likely to be younger, female, anhedonic, depressed, and anxious. In both Classes 3 and 4, functional impairment, physical symptom burden, and suicidal ideation were present. Conclusion(s): In contrast with the severe symptom cluster carrying anhedonia, anxiety, and demoralization, the moderate symptom cluster was formed by patients with demoralization and impaired functioning, a clinical picture consistent with a unidimensional model of adjustment disorder.Copyright © 2018 John Wiley & Sons, Ltd.

Background: A systematic review suggests that 25% of oesophageal adenocarcinomas (OAC) are 'missed' at index endoscopy for Barrett's oesophagus (BO); however, this included few population-based studies and may be an overestimate. Objective(s): The objective of this article is to quantify the 'missed' rates of high-grade dysplasia (HGD) and OAC at index BO endoscopy. Method(s): Patients from the Northern Ireland BO register diagnosed between 1993 and 2010 (n = 13,159) were linked to the Northern Ireland Cancer Registry to identify patients who developed OAC or HGD. Logistic regression analysis compared characteristics of 'missed' vs 'incident' HGD/OAC, defined as diagnoses within 3-12 months vs >1 year after incident BO, respectively. Result(s): A total of 267 patients were diagnosed with HGD/OAC >=3 months after BO diagnosis, of whom 34 (12.7%) were potentially 'missed'. The proportion of 'missed' HGD/OAC was 25% among BO patients with low-grade dysplasia (LGD) and 9% among non-dysplastic BO patients. Older age and BO-LGD carried a higher risk of 'missed' HGD/OAC. Non-dysplastic BO patients were more often diagnosed with a 'missed' OAC (rather than HGD; 89%), compared with BO-LGD patients (40%). Conclusion(s): Approximately one in 10 HGD/OAC cases are 'missed' at incident BO diagnosis, which is significant but lower than previous reports. However, 'missed' HGD/OAC cases represent only 0.26% of all BO patients.Copyright © 2017, © Author(s) 2017.

Full Text, Thesis (PhD Doctorate), Doctor of Philosophy (PhD), School of Crim & Crim Justice, Arts, Education and Law, BACKGROUND Persistent offending has been a topic of interest in the criminal justice literature for some time. Despite its popularity, the concept remains without a consistent or agreed upon definition. For example, persistent offending has been variably defined as offending before and during adulthood (e.g., Bergman & Andershed, 2009), frequent offending (e.g., McGloin & Stickle, 2011), or an early age of first offence (e.g. Hay & Forrest, 2009). In turn, inconsistent methods have been used to operationalise and identify persistent offenders, such as offending before and after the age of 21 (Farrington, Ttofi, & Coid, 2009), the five percent most frequent offenders (Piquero & Lawton, 2002), and multiple offences committed before the age of 14 (Hagell & Newburn, 1994). The definitional inconsistency surrounding the concept of persistent offending poses a significant threat to the generalisability of research, accuracy of theory, and efficacy of policy and interventions. Although previous authors have highlighted their concerns regarding the methodological inconsistencies pertaining to research on persistent offending (Hagell & Newburn, 1994; Piquero, 2009), these concerns seem to have fallen on deaf ears. AIMS This dissertation argues that the use of inconsistent definitions and operationalisations of persistent offending are contributing to the inconsistent findings and competing explanations on the phenomenon. Therefore, more conceptual discussions and empirical observations drawing attention to the ramifications of this issue, as well as methods for rectifying the problem, are needed. Through a series of published and unpublished papers, this dissertation attempts to meet this need by: (1) Highlighting the prevalence of inconsistent definitions, operationalisations, and measures of persistent offending in the literature, and the consequent need for consistency; (2) Empirically demonstrating the flaws associated with these inconsistencies, and; (3) Proposing how t

Fixed-object crashes have long time been considered as major roadway safety concerns. While previous relevant studies tended to address such crashes in the context of roadway departures, and heavily relied on police-reported accidents data, this study integrated the SHRP2 NDS and RID data for analyses, which fully depicted the prior to, during, and after crash scenarios. A total of 1,639 crash, near-crash events, and 1,050 baseline events were acquired. Three analysis methods: logistic regression, support vector machine (SVM) and artificial neural network (ANN) were employed for two responses: crash occurrence and severity level. Logistic regression analyses identified 16 and 10 significant variables with significance levels of 0.1, relevant to driver, roadway, environment, etc. for two responses respectively. The logistic regression analyses led to a series of findings regarding the effects of explanatory variables on fixed-object event occurrence and associated severity level. SVM classifiers and ANN models were also constructed to predict these two responses. Sensitivity analyses were performed for SVM classifiers to infer the contributing effects of input variables. All three methods obtained satisfactory prediction performance, that was around 88% for fixed-object event occurrence and 75% for event severity level, which indicated the effectiveness of NDS event data on depicting crash scenarios and roadway safety analyses.

El microcrédito puede ser visto como un conjunto de operaciones definidas dentro de la ley y reguladas por el sistema financiero, que se destinan a realizar préstamos de monto reducido con condiciones especiales, en tasa de interés y plazos de amortización a microempresarios, hogares y personas que necesiten inversión en capital de trabajo, compra de activos fijos o de cartera en sus proyectos productivos de etapa inicial o andantes -- Teniendo en cuenta que esta fuente de financiación posee un gran potencial de desarrollo, dada la cantidad de microempresas e informalismo en el país, y que desde el año 2011 ha sido la cartera con mayor participación en cuanto al número de deudores, pues viene creciendo a un ritmo superior que la cartera total del sistema financiero además de presentar un deterioro en los indicadores de riesgo de crédito, el objetivo del presente trabajo es realizar una medición de los determinantes cualitativos del incumplimiento de pago de la cartera del microcrédito en Colombia -- Para ello, se hace uso del concepto de información suave, la cual incluye variables socioeconómicas y características de las microempresas y los deudores que pueden determinar el impacto de las mismas sobre la probabilidad de mora -- Usando la metodología modelo de regresión logística se obtendrán evidencias estadísticas a favor o en contra de la hipótesis, The microcredit can be seen as a set of operations defined within the law and regulated by the financial system that are intended to make loans of reduced amount with special conditions in interest rates and amortization periods to microentrepreneurs, households and people who need investment in working capital, purchase of fixed assets or portfolio in their productive projects of initial stage and/ or errant -- Bearing in mind that this source of financing has a great potential for development given the number of micro-enterprises and informalism in the country and that since 2011 has been the portfolio with the largest share in the number of debtors, as it has been growing at a higher than the total portfolio of the financial system -- In addition to presenting a deterioration in the credit risk indicators, the qualitative determinants of non-payment of the microcredit portfolio in Colombia -- To do this, the concept of soft information is used, which includes socioeconomic variables and characteristics of microenterprises and debtors that can determine their impact on the probability of default -- Using the logistic regression model methodology, you will obtain statistical evidence for or against the hypothesis

Los modelos de inteligencia artificial son un problema abierto para aplicación en diversos campos de la ciencia y la identificación de relaciones entre variables, en especial cuando la distribución de los sucesos no depende de una función lineal -- Al considerar lo anterior, esta investigación busca establecer una comparación entre un método tradicional utilizado para el seguimiento del comportamiento crediticio y modelos avanzados de inteligencia artificial -- Las guías que en la actualidad existen en el país y que están dadas por la Superintendencia Financiera de Colombia para el manejo del riesgo de crédito, así como los estándares internacionales, como Basilea II, Basilea III y Solvency, están basados en regresiones logísticas y en análisis discriminantes, modelos que usan las entidades financieras para medir el comportamiento crediticio, por lo que se propone explorar la utilidad de nuevas metodologías -- En este trabajo se aborda uno de los métodos tradicionales utilizados en las instituciones financieras, esto es, la regresión logística, y se compara con métodos alternativos, como las redes neuronales y los bosques aleatorios -- A partir de la revisión de literatura y mediante la utilización de una base de datos suministrada por una entidad bancaria, se seleccionaron las variables dependientes y la variable de respuesta, se calibraron los modelos de regresión logística, lo mismo que bosques aleatorios y redes neuronales en el aplicativo Microsoft Azure Machine Learning y se compararon entre sí con indicadores de precisión y exactitud, como la curva de tipo ROC (por las iniciales de la expresión en inglés receiver operating characteristic) y la matriz de confusión; se obtuvieron para los modelos de inteligencia artificial resultados tan buenos como el tradicional, por lo que pueden ser utilizados por el sector financiero como métodos alternos o complementarios en el análisis de riesgo de crédito, Artificial intelligence models are an open problem for application in various fields of science and search of variable relationships especially when the distribution of events doesn’t depend on a linear function; through this work we want to compare the traditional method most used for credit behavior monitoring with advanced models of artificial intelligence -- The guides that exist in Colombia for management of credit risk are given by the Financial Superintendence of Colombia, international standards such as Basel II, Basel III and Solvency are based on the logistic regression and the discriminant analysis, models used by financial institutions in Colombia to measure credit behavior, thus we carried out an investigation to explore the utility of new models -- This paper addresses one of the traditional methods used in financial institutions, that is, logistic regression, and compares it with alternative methods such as neural networks and random forests -- From the literature review and using a database provided by a banking entity, the dependent variables and the response variable are selected, the logistic regression models, random forests and neural networks are calibrated in the Microsoft Azure Machine Learning application and they are compared to each other with indicators of precision and accuracy such as ROC (from receiver operating characteristic) curve and confusion matrix, obtaining for the models of artificial intelligence, results as good as the traditional one; so they can be used by the financial sector as alternate and / or complementary methods in the analysis of credit risk

EMEESA es una empresa generadora, distribuidora y comercializadora de energía en el mercado eléctrico colombiano, la cual brinda el servicio de energía a diferentes poblaciones del departamento del Cauca -- Las condiciones socioeconómicas de éste han elevado la frecuencia de impago a la empresa -- En este sentido, en el presente trabajo se establecen los factores que determinan esta condición, lo que sirve de base para la generación de estrategias que fortalezcan el esquema comercial a partir de la información disponible -- Se propone así la elaboración de un modelo logístico para estimar la probabilidad de incumplimiento de cada cliente y, a través del modelo Montecarlo, obtener una aproximación de las pérdidas esperadas para la compañía en lo que a cartera morosa se refiere, frente a lo cual se encuentra que el rango de consumo de energía aumenta la probabilidad de incumplimiento, EMESSA is a generator, distributor and marketing company of energy from the Colombian electric market which gives the energy services for some populated downtowns of Cauca’s department -- The socioeconomic conditions of this department had increase the unpaynment frequency to the company -- In this work, the factors which are established determines this condition -- This would serve as a base for the generation of new strategies that would make an stronger comertial scheme -- Through the avaliable information, it is proposed the estimation of a logistic model for estimate the unpaynment of Each user, and posteriorly, trough the Montecarlos logistic model we obtain an approximation of the waited losses for the company in what delayed pursue refers -- Its found that the consume range increases the unpaynment probability

Las empresas de servicios financieros son parte fundamental en la economía del país, y su continuo crecimiento ha ocasionado que sean más susceptibles al fraude debido a la falta de prevención y de controles que mitiguen las amenazas que van de la mano con la innovación tecnológica, los nuevos productos y las características del consumo -- Esta problemática genera impactos negativos a nivel económico y social tanto para las empresas como para el país -- Es por esto que se identificó la necesidad de contribuir a la elaboración del perfil de riesgo del defraudador interno en una entidad financiera que opera en Colombia, para ayudar en la prevención y detección de fraudes -- Así, se definió un perfil parcial (puesto que solo contempla algunas variables) a partir del uso de un modelo estadístico predictivo (con funciones más probabilísticas que predictivas en el caso del fraude) que analizó la información proveniente de la caracterización de los empleados implicados en acciones fraudulentas, además de una revisión de literatura -- A partir de la construcción del modelo de Random Forest, y con el objetivo de encontrar empleados con altas probabilidades de incurrir en actos incorrectos, se identificó como perfil defraudador de la entidad, los trabajadores que laboran en la red de sucursales, que poseen transacciones atípicas y en donde las relaciones de salarios vs los descuentos realizados al empleado no guardan una proporción -- Los empleados que poseen cargos operativos, los cuales representan el 44.5 % de la empresa, son los más propensos a incurrir en fraude interno, el 66.4 % de los posibles defraudadores son mujeres; y las regiones del país que presentan mayor alerta son la región centro, con el 43.6 %, seguida de Bogotá y Sabana con el 28.7%, Financial services companies are a fundamental part of the country's economy, and their continued growth has made them more susceptible to fraud due to the lack of prevention and controls that mitigate the threats that go hand in hand with technological innovation, new products and consumption characteristics -- This problem generates negative economic and social impacts for both the company and the country -- That is why it was identified the need to develop the risk profile of the internal fraudster in a financial entity, which allows the frauds prevention and detection from the characterization of employees involved in fraudulent actions, to achieve this, the mixed investigation was applied, which allowed the collection of data on fraud materialized by employees, their numerical measurement and respective statistical analysis to test the hypothesis, as well as a literature review -- From the construction of the Random Forest model, and with the objective of finding employees with high probabilities of incurring in incorrect acts, the profile of the internal fraudster of the entity was identified, the employee who work in the branch network, who have atypical transactions and where the wage relationships Vs the discounts made to the employee do not keep a proportion -- Employees who hold operational positions which represent the 44.5% of the company, are the most likely to incur internal fraud, 66.4% of possible fraudsters are women; moreover, country’s regions that present greater alert are the central region with 43.6%, followed by Bogotá and Sabana with 28.7%