Search results for: binary logistic regression

Authors: Thoetchai Peeraphatdit, Peter A. Brady, Suraj Kapa, Samuel J. Asirvatham, Niyada Naksuk

Abstract:

Background: Serum electrolyte abnormalities are a common cause of an acquired prolonged QT syndrome, especially, in the coronary care unit (CCU) setting. Optimal electrolyte ranges among the CCU patients have not been sufficiently investigated. Methods: We identified 8,498 consecutive CCU patients who were admitted to the CCU at Mayo Clinic, Rochester, the USA, from 2004 through 2013. Association between first serum electrolytes and baseline corrected QT intervals (QTc), as well as in-hospital mortality, was tested using multivariate linear regression and logistic regression, respectively. Serum potassium 4.0- < 4.5 mEq/L, ionized calcium (iCa) 4.6-4.8 mg/dL, and magnesium 2.0- < 2.2 mg/dL were used as the reference levels. Results: There was a modest level-dependent relationship between hypokalemia ( < 4.0 mEq/L), hypocalcemia ( < 4.4 mg/dL), and a prolonged QTc interval; serum magnesium did not affect the QTc interval. Association between the serum electrolytes and in-hospital mortality included a U-shaped relationship for serum potassium (adjusted odds ratio (OR) 1.53 and OR 1.91for serum potassium 4.5- < 5.0 and ≥ 5.0 mEq/L, respectively) and an inverted J-shaped relationship for iCa (adjusted OR 2.79 and OR 2.03 for calcium < 4.4 and 4.4- < 4.6 mg/dL, respectively). For serum magnesium, the mortality was greater only among patients with levels ≥ 2.4 mg/dL (adjusted OR 1.40), compared to the reference level. Findings were similar in sensitivity analyses examining the association between mean serum electrolytes and mean QTc intervals, as well as in-hospital mortality. Conclusions: Serum potassium 4.0- < 4.5 mEq/L, iCa ≥ 4.6 mg/dL, and magnesium < 2.4 mg/dL had a neutral effect on QTc intervals and were associated with the lowest in-hospital mortality among the CCU patients.

Keywords: calcium, electrocardiography, long-QT syndrome, magnesium, mortality, potassium

Procedia PDF Downloads 394

Authors: Matthew Cardeiro, Amalia D. Ardeljan, Lexi Frankel, Dianela Prado Escobar, Catalina Molnar, Omar M. Rashid

Abstract:

Introduction: Enterococci comprise the natural flora of nearly all animals and are ubiquitous in food manufacturing and probiotics. However, its role in the microbiome remains controversial. The gut microbiome has shown to play an important role in immunology and cancer. Further, recent data has suggested a relationship between gut microbiota and breast cancer. These studies have shown that the gut microbiome of patients with breast cancer differs from that of healthy patients. Research regarding enterococcus infection and its sequala is limited, and further research is needed in order to understand the relationship between infection and cancer. Enterococcus may prevent the development of breast cancer (BC) through complex immunologic and microbiotic adaptations following an enterococcus infection. This study investigated the effect of enterococcus infection and the incidence of BC. Methods: A retrospective study (January 2010- December 2019) was provided by a Health Insurance Portability and Accountability Act (HIPAA) compliant national database and conducted using a Humans Health Insurance Database. International Classification of Disease (ICD) 9th and 10th codes, Current Procedural Terminology (CPT), and National Drug Codes were used to identify BC diagnosis and enterococcus infection. Patients were matched for age, sex, Charlson Comorbidity Index (CCI), antibiotic treatment, and region of residence. Chi-squared, logistic regression, and odds ratio were implemented to assess the significance and estimate relative risk. Results: 671 out of 28,518 (2.35%) patients with a prior enterococcus infection and 1,459 out of 28,518 (5.12%) patients without enterococcus infection subsequently developed BC, and the difference was statistically significant (p<2.2x10⁻¹⁶). Logistic regression also indicated enterococcus infection was associated with a decreased incidence of BC (RR=0.60, 95% CI [0.57, 0.63]). Treatment for enterococcus infection was analyzed and controlled for in both enterococcus infected and noninfected populations. 398 out of 11,523 (3.34%) patients with a prior enterococcus infection and treated with antibiotics were compared to 624 out of 11,523 (5.41%) patients with no history of enterococcus infection (control) and received antibiotic treatment. Both populations subsequently developed BC. Results remained statistically significant (p<2.2x10-16) with a relative risk of 0.57 (95% CI [0.54, 0.60]). Conclusion & Discussion: This study shows a statistically significant correlation between enterococcus infection and a decrease incidence of breast cancer. Further exploration is needed to identify and understand not only the role of enterococcus in the microbiome but also the protective mechanism(s) and impact enterococcus infection may have on breast cancer development. Ultimately, further research is needed in order to understand the complex and intricate relationship between the microbiome, immunology, bacterial infections, and carcinogenesis.

Keywords: breast cancer, enterococcus, immunology, infection, microbiome

Procedia PDF Downloads 173

Authors: Zikiou Nadia, Lahdir Mourad, Ameur Soltane

Abstract:

In this paper, we propose an effective method for image compression based on SVM Regression (SVR), with three different kernels, and biorthogonal 2D Discrete Wavelet Transform. SVM regression could learn dependency from training data and compressed using fewer training points (support vectors) to represent the original data and eliminate the redundancy. Biorthogonal wavelet has been used to transform the image and the coefficients acquired are then trained with different kernels SVM (Gaussian, Polynomial, and Linear). Run-length and Arithmetic coders are used to encode the support vectors and its corresponding weights, obtained from the SVM regression. The peak signal noise ratio (PSNR) and their compression ratios of several test images, compressed with our algorithm, with different kernels are presented. Compared with other kernels, Gaussian kernel achieves better image quality. Experimental results show that the compression performance of our method gains much improvement.

Keywords: image compression, 2D discrete wavelet transform (DWT-2D), support vector regression (SVR), SVM Kernels, run-length, arithmetic coding

Procedia PDF Downloads 382

Authors: H. Hruschka

Abstract:

This paper presents the first comparison of the performance of the restricted Boltzmann machine and the deep belief net on binary market basket data relative to binary factor analysis and the two best-known topic models, namely Dirichlet allocation and the correlated topic model. This comparison shows that the restricted Boltzmann machine and the deep belief net are superior to both binary factor analysis and topic models. Managerial implications that differ between the investigated models are treated as well. The restricted Boltzmann machine is defined as joint Boltzmann distribution of hidden variables and observed variables (purchases). It comprises one layer of observed variables and one layer of hidden variables. Note that variables of the same layer are not connected. The comparison also includes deep belief nets with three layers. The first layer is a restricted Boltzmann machine based on category purchases. Hidden variables of the first layer are used as input variables by the second-layer restricted Boltzmann machine which then generates second-layer hidden variables. Finally, in the third layer hidden variables are related to purchases. A public data set is analyzed which contains one month of real-world point-of-sale transactions in a typical local grocery outlet. It consists of 9,835 market baskets referring to 169 product categories. This data set is randomly split into two halves. One half is used for estimation, the other serves as holdout data. Each model is evaluated by the log likelihood for the holdout data. Performance of the topic models is disappointing as the holdout log likelihood of the correlated topic model – which is better than Dirichlet allocation - is lower by more than 25,000 compared to the best binary factor analysis model. On the other hand, binary factor analysis on its own is clearly surpassed by both the restricted Boltzmann machine and the deep belief net whose holdout log likelihoods are higher by more than 23,000. Overall, the deep belief net performs best. We also interpret hidden variables discovered by binary factor analysis, the restricted Boltzmann machine and the deep belief net. Hidden variables characterized by the product categories to which they are related differ strongly between these three models. To derive managerial implications we assess the effect of promoting each category on total basket size, i.e., the number of purchased product categories, due to each category's interdependence with all the other categories. The investigated models lead to very different implications as they disagree about which categories are associated with higher basket size increases due to a promotion. Of course, recommendations based on better performing models should be preferred. The impressive performance advantages of the restricted Boltzmann machine and the deep belief net suggest continuing research by appropriate extensions. To include predictors, especially marketing variables such as price, seems to be an obvious next step. It might also be feasible to take a more detailed perspective by considering purchases of brands instead of purchases of product categories.

Keywords: binary factor analysis, deep belief net, market basket analysis, restricted Boltzmann machine, topic models

Procedia PDF Downloads 199

Authors: Adriano Z. Zambom, Preethi Ravikumar

Abstract:

One of the biggest challenges in nonparametric regression is the curse of dimensionality. Additive models are known to overcome this problem by estimating only the individual additive effects of each covariate. However, if the model is misspecified, the accuracy of the estimator compared to the fully nonparametric one is unknown. In this work the efficiency of completely nonparametric regression estimators such as the Loess is compared to the estimators that assume additivity in several situations, including additive and non-additive regression scenarios. The comparison is done by computing the oracle mean square error of the estimators with regards to the true nonparametric regression function. Then, a backward elimination selection procedure based on the Akaike Information Criteria is proposed, which is computed from either the additive or the nonparametric model. Simulations show that if the additive model is misspecified, the percentage of time it fails to select important variables can be higher than that of the fully nonparametric approach. A dimension reduction step is included when nonparametric estimator cannot be computed due to the curse of dimensionality. Finally, the Boston housing dataset is analyzed using the proposed backward elimination procedure and the selected variables are identified.

Keywords: additive model, nonparametric regression, variable selection, Akaike Information Criteria

Procedia PDF Downloads 265

Authors: Masood Beheshtirad

Abstract:

Identification of regions having potential for landslide occurrence is one of the basic measures in natural resources management. Different landslide hazard mapping models are proposed based on the environmental condition and goals. In this research landslide hazard map using multiple regression model were provided and applicability of this model is investigated in Baghdasht watershed. Dependent variable is landslide inventory map and independent variables consist of information layers as Geology, slope, aspect, distance from river, distance from road, fault and land use. For doing this, existing landslides have been identified and an inventory map made. The landslide hazard map is based on the multiple regression provided. The level of similarity potential hazard classes and figures of this model were compared with the landslide inventory map in the SPSS environments. Results of research showed that there is a significant correlation between the potential hazard classes and figures with area of the landslides. The multiple regression model is suitable for application in the Baghdasht Watershed.

Keywords: landslide, mapping, multiple model, regression

Procedia PDF Downloads 325

Authors: Fouzia Amireche-Ziar, Ilham Mokbel, Jacques Jose

Abstract:

The vapour pressures of the three binary mixtures: 1, 2- dichloroethane + 1,3-dioxolane, + 1,4-dioxane or + tetrahydropyrane, are carried out at ten temperatures ranging from 273 to 353.15 K. An accurate static device was employed for these measurements. The VLE data were reduced using the Redlich-Kister equation by taking into consideration the vapour pressure non-ideality in terms of the second molar virial coefficient. The experimental data were compared to the results predicted with the DISQUAC and Dortmund UNIFAC group contribution models for the total pressures P and the excess molar Gibbs energies GE.

Keywords: disquac model, dortmund UNIFAC model, excess molar Gibbs energies GE, VLE

Procedia PDF Downloads 228

Authors: Nora Gavira Duron, Cinthya Moreda Gonzalez-Ortega, Reyna Susana Garcia Ruiz

Abstract:

This paper objective is to analyze the mathematics performance in the Learning Evaluation National Plan (PLANEA for its Spanish initials: Plan Nacional para la Evaluación de los Aprendizajes), applied to Mexican students who are enrolled in the last elementary-school year over the 2017-2018 academic year. Such test was conducted nationwide in 3,573 schools, using a sample of 108,083 students, whose average in mathematics, on a scale of 0 to 100, was 45.6 points. 75% of the sample analyzed did not reach the sufficiency level (60 points). It should be noted that only 2% got a 90 or higher score result. The performance is analyzed while considering whether there are differences in gender, marginalization level, public or private school enrollment, parents’ academic background, and living-with-parents situation. Likewise, this variable impact (among other variables) on school performance by gender is evaluated, considering multivariate logistic (Logit) regression analysis. The results show there are no significant differences in mathematics performance regarding gender in elementary school; nevertheless, the impact exerted by mothers who studied at least high school is of great relevance for students, particularly for girls. Other determining variables are students’ resilience, their parents’ economic status, and the fact they attend private schools, strengthened by the mother's education.

Keywords: multivariate regression analysis, academic performance, learning evaluation, mathematics result per gender

Procedia PDF Downloads 147

Authors: Rakesh Kumar Mishra, Tarun Kumar Dan

Abstract:

Lead-Lag based Internal Model Control method is proposed based on Internal Model Control (IMC) strategy. In this paper, we have designed the Lead-Lag based Internal Model Control for binary distillation column for SISO process (considering only bottom product). The transfer function has been taken from Wood and Berry model. We have find the composition control and disturbance rejection using Lead-Lag based IMC and comparing with the response of simple Internal Model Controller.

Keywords: SISO, lead-lag, internal model control, wood and berry, distillation column

Procedia PDF Downloads 646

Authors: Ebenezer Kwesi Armah-Ansah, Kenneth Fosu Oteng, Esther Selasi Avinu, Eugene Budu, Edward Kwabena Ameyaw

Abstract:

Background: Maternal mortality, particularly in Benin, is a major public health concern in Sub-Saharan Africa. To provide a positive pregnancy experience and reduce maternal morbidities, all pregnant women must get appropriate and timely prenatal support. However, many pregnant women in developing countries, including Benin, begin antenatal care late. There is a paucity of empirical literature on the prevalence and predictors of early antenatal care visits in Benin. As a result, the purpose of this study is to investigate the prevalence and predictors of early antenatal care visits among women of productive age in Benin. Methods: This is a secondary analysis of the 2017/2018 Benin Demographic and Health Survey (BDHS) data. The study involved 6,919 eligible women. Data analysis was conducted using Stata version 14.2 for Mac OS. We adopted a multilevel logistic regression to examine the predictors of early ANC visits in Benin. The results were presented as odds ratios (ORs) associated with 95% confidence intervals (CIs) and p-value <0.05 to determine the significant associations. Results: The prevalence of early ANC visits among pregnant women in Benin was 57.03% [95% CI: 55.41-58.64]. In the final multilevel logistic regression, early ANC visit was higher among women aged 30-34 [aOR=1.60, 95% CI=1.17-2.18] compared to those aged 15-19, women with primary education [aOR=1.22, 95% CI=1.06-142] compared to the non-educated women, women who were covered by health insurance [aOR=3.03, 95% CI=1.35-6.76], women without a big problem in getting the money needed for treatment [aOR=1.31, 95% CI=1.16-1.49], distance to the health facility, not a big problem [aOR=1.23, 95% CI=1.08-1.41], and women whose partners had secondary/higher education [aOR=1.35, 95% CI=1.15-1.57] compared with those who were not covered by health insurance, had big problem in getting money needed for treatment, distance to health facility is a big problem and whose partners had no education respectively. However, women who had four or more births [aOR=0.60, 95% CI=0.48-0.74] and those in Atacora Region [aOR=0.50, 95% CI=0.37-0.68] had lower odds of early ANC visit. Conclusion: This study revealed a relatively high prevalence of early ANC visits among women of reproductive age in Benin. Women's age, educational status of women and their partners, parity, health insurance coverage, distance to health facilities, and region were all associated with early ANC visits among women of reproductive in Benin. These factors ought to be taken into account when developing ANC policies and strategies in order to boost early ANC visits among women in Benin. This will significantly reduce maternal and newborn mortality and help achieve the World Health Organization’s recommendation that all pregnant women should initiate early ANC visits within the first three months of pregnancy.

Keywords: antenatal care, Benin, maternal health, pregnancy, DHS, public health

Procedia PDF Downloads 66

Authors: Helena Baffoe

Abstract:

The US has developed numerous programs over the past 50 years to enhance the lives of those who suffer from mental health illnesses and substance abuse, as well as the effectiveness of their treatments. Despite these advances over the past 50 years, there hasn't been a corresponding improvement in American public attitudes and beliefs about mental health disorders and substance abuse. Highly publicized acts of violence frequently elicit comments that blame the perpetrator's perceived mental health disorder since such people are thought to be substance abusers. Despite these strong public beliefs and perception about mental disorder and substance abuse, concreate empirical evidence that entail this perception is lacking, and evidence of their effectiveness has not been integrated. A rich data was collected from Substance Abuse and Mental Health Services Administration (SAMHSA) with a hypothesis that people who are diagnosed with a mental health disorder are likely to be diagnosed with substance abuse using logit regression analysis and Instrumental Variable. It was found that depressive, anxiety, and trauma/stressor mental disorders constitute the most common mental disorder in the United States, and the study could not find statistically significant evidence that being diagnosed with these leading mental health disorders in the United States does necessarily imply that such a patient is diagnosed with substances abuse. Thus, the public has a misconception of mental health and substance abuse issues, and social workers' responsibilities are outlined in order to assist ameliorate this attitude and perception.

Keywords: mental health disorder, substance use, empirical evidence, logistic regression

Procedia PDF Downloads 78

Authors: Syrus Zimaze, George Nsimbe, Valley Mugwanya, Matiya Lule, Edgar Watson, Patrick Gwayambadde

Abstract:

Physical inactivity is a growing global epidemic, also recognised as a major public health challenge. Globally, there are alarming rates of children reported with cardiovascular disease and obesity with limited interventions. In Sub Saharan Africa, there is limited information about involvement in physical activity especially among children aged 6 to 18 years. The aim of this study was to explore factors associated with involvement in physical activity among children in Uganda. Methods: We included all parents with children aged 6 to 18 years training with Excel Soccer Academy between January 2017 and June 2018. Physical activity definition was time spent participating in routine soccer training at the academy for more than 30 days. Each child's attendance was recorded, and parents provided demographic and social economic data. Data on predictors of physical activity involvement were collected using a standardized questionnaire. Descriptive statistics and frequency were used. Binary logistic regression was used at the multi variable level adjusting for education, residence, transport means and access to information technology. Results: Overall 356 parents were interviewed; Boys 318 (89.3%) engaged more in physical activity than girls. The median age for children was 13 years (IQR:6-18) and 42 years (IQR:37-49) among parents. The median time spent at the Excel soccer academy was 13.4 months (IQR: 4.6-35.7) Majority of the children attended formal education, p < 0.001). Factors associated with involvement in physical activity included: owning a permanent house compared to a rented house (odds ratio [OR] :2.84: 95% CI: 2.09-3.86, p < 0.0001), owning a car compared to using public transport (OR: 5.64 CI: 4.80-6.63, p < 0.0001), a parent having received formal education compared to non-formal education (OR: 2.93 CI: 2.47-3.46, p < 0.0001) and daily access to information technology (OR:0.40 CI:0.25-0.66, p < 0.001). Parent’s age and gender were not associated to involvement in physical activity. Conclusions: Socioeconomic factors were positively associated with involvement in physical activity with boys participating more than girls in soccer activities. More interventions are required geared towards increasing girl’s participation in physical activity and those targeting children from less privilege homes.

Keywords: physical activity, Sub-Saharan Africa, social economic factors, children

Procedia PDF Downloads 166

Authors: S. Chami, J. Chauvin, T. Demarest, Stan Ng, M. Straus, W. Jahner

Abstract:

Biometrics tools such as fingerprint and iris are widely used in industry to protect critical assets. However, their vulnerability and lack of robustness raise several worries about the protection of highly critical assets. Biometrics based on Electrocardiographic (ECG) signals is a robust identification tool. However, most of the state-of-the-art techniques have worked on clinical signals, which are of high quality and less noisy, extracted from wearable devices like a smartwatch. In this paper, we are presenting a complete machine learning pipeline that identifies people using ECG extracted from an off-person device. An off-person device is a wearable device that is not used in a medical context such as a smartwatch. In addition, one of the main challenges of ECG biometrics is the variability of the ECG of different persons and different situations. To solve this issue, we proposed two different approaches: per person classifier, and one-for-all classifier. The first approach suggests making binary classifier to distinguish one person from others. The second approach suggests a multi-classifier that distinguishes the selected set of individuals from non-selected individuals (others). The preliminary results, the binary classifier obtained a performance 90% in terms of accuracy within a balanced data. The second approach has reported a log loss of 0.05 as a multi-class score.

Keywords: biometrics, electrocardiographic, machine learning, signals processing

Procedia PDF Downloads 142

Authors: Javad Izadi, Sahar Majioud

Abstract:

This study presents an approach of assessing the choice of performance measures of companies in the United Kingdom after the application of IFRS in 2005. The aim of this study is to investigate the effects of IFRS on the choice of performance evaluation methods for UK companies. We analyse through an econometric model the relationship of the dependent variable, the firm’s performance, which is a nominal variable with the independent ones. Independent variables are split into two main groups: the first one is the group of accounting-based measures: Earning per share, return on assets and return on equities. The second one is the group of market-based measures: market value of property plant and equipment, research and development, sales growth, market to book value, leverage, segment and size of companies. Concerning the regression used, it is a multinomial logistic regression performed on a sample of 130 UK listed companies. Our finding shows after IFRS adoption, and companies give more importance to some variables such as return on equities and sales growth to assess their performance, whereas the return on assets and market to book value ratio does not have as much importance as before IFRS in evaluating the performance of companies. Also, there are some variables that have no impact on the performance measures anymore, such as earning per share. This article finding is empirically important for business in subjects related to IFRS and companies’ performance measurement.

Keywords: performance’s Measure, nominal variable, econometric model, evaluation methods

Procedia PDF Downloads 138

Authors: Arun Goel

Abstract:

Prediction of maximum local scour is necessary for the safety and economical design of the bridges. A number of equations have been developed over the years to predict local scour depth using laboratory data and a few pier equations have also been proposed using field data. Most of these equations are empirical in nature as indicated by the past publications. In this paper, attempts have been made to compute local depth of scour around bridge pier in dimensional and non-dimensional form by using linear regression, simple regression and SVM (Poly and Rbf) techniques along with few conventional empirical equations. The outcome of this study suggests that the SVM (Poly and Rbf) based modeling can be employed as an alternate to linear regression, simple regression and the conventional empirical equations in predicting scour depth of bridge piers. The results of present study on the basis of non-dimensional form of bridge pier scour indicates the improvement in the performance of SVM (Poly and Rbf) in comparison to dimensional form of scour.

Keywords: modeling, pier scour, regression, prediction, SVM (Poly and Rbf kernels)

Procedia PDF Downloads 451

Authors: Senol Dogan, Gunay Karli

Abstract:

Identification of relevant publications based on a Medline query is time-consuming and error-prone. An all based process has the potential to solve this problem without any manual work. To the best of our knowledge, our study is the first to investigate the ability of machine learning to identify relevant articles accurately. 5 different machine learning algorithms were tested using 23 predictors based on several metadata fields attached to publications. We find that the Boosted model is the best-performing algorithm and its overall accuracy is 96%. In addition, specificity and sensitivity of the algorithm is 97 and 93%, respectively. As a result of the work, we understood that we can apply the same procedure to understand cancer gene expression big data.

Keywords: prediction of publications, machine learning, algorithms, bariatric surgery, comparison of algorithms, boosted, tree, logistic regression, ANN model

Procedia PDF Downloads 210

Authors: Abdullah A. AlShaher

Abstract:

In this paper, we demonstrate how regression curves can be used to recognize 2D non-rigid handwritten shapes. Each shape is represented by a set of non-overlapping uniformly distributed landmarks. The underlying models utilize 2^nd order of polynomials to model shapes within a training set. To estimate the regression models, we need to extract the required coefficients which describe the variations for a set of shape class. Hence, a least square method is used to estimate such modes. We then proceed by training these coefficients using the apparatus Expectation Maximization algorithm. Recognition is carried out by finding the least error landmarks displacement with respect to the model curves. Handwritten isolated Arabic characters are used to evaluate our approach.

Keywords: character recognition, regression curves, handwritten Arabic letters, expectation maximization algorithm

Procedia PDF Downloads 145

Authors: Khandoker Akib Mohammad

Abstract:

While analyzing time-to-event data, it is possible that a certain fraction of subjects will never experience the event of interest, and they are said to be cured. When this feature of survival models is taken into account, the models are commonly referred to as cure models. In the presence of covariates, the conditional survival function of the population can be modelled by using the cure model, which depends on the probability of being uncured (incidence) and the conditional survival function of the uncured subjects (latency), and a combination of logistic regression and Cox proportional hazards (PH) regression is used to model the incidence and latency respectively. In this paper, we have shown the asymptotic normality of the profile likelihood estimator via asymptotic expansion of the profile likelihood and obtain the explicit form of the variance estimator with an implicit function in the profile likelihood. We have also shown the efficient score function based on projection theory and the profile likelihood score function are equal. Our contribution in this paper is that we have expressed the efficient information matrix as the variance of the profile likelihood score function. A simulation study suggests that the estimated standard errors from bootstrap samples (SMCURE package) and the profile likelihood score function (our approach) are providing similar and comparable results. The numerical result of our proposed method is also shown by using the melanoma data from SMCURE R-package, and we compare the results with the output obtained from the SMCURE package.

Keywords: Cox PH model, cure model, efficient score function, EM algorithm, implicit function, profile likelihood

Procedia PDF Downloads 144

Authors: Dalia Rimawi, Walid Salameh, Amal Al-Omari, Hadeel AbdelKhaleq

Abstract:

Predication of Survival time of patients with cancer, is a core factor that influences oncologist decisions in different aspects; such as offered treatment plans, patients’ quality of life and medications development. For a long time proportional hazards Cox regression (ph. Cox) was and still the most well-known statistical method to predict survival outcome. But due to the revolution of data sciences; new predication models were employed and proved to be more flexible and provided higher accuracy in that type of studies. Artificial neural network is one of those models that is suitable to handle time to event predication. In this study we aim to compare ph Cox regression with artificial neural network method according to data handling and Accuracy of each model.

Keywords: Cox regression, neural networks, survival, cancer.

Procedia PDF Downloads 200

Authors: Al Omari Mohammed Ahmed

Abstract:

This paper focuses on Maximum Likelihood Estimator with Covariate. Covariates are incorporated into the Weibull model. Under this regression model with regards to maximum likelihood estimator, the parameters of the covariate, shape parameter, survival function and hazard rate of the Weibull regression distribution with right censored data are estimated. The mean square error (MSE) and absolute bias are used to compare the performance of Weibull regression distribution. For the simulation comparison, the study used various sample sizes and several specific values of the Weibull shape parameter.

Keywords: weibull regression distribution, maximum likelihood estimator, survival function, hazard rate, right censoring

Procedia PDF Downloads 441

Authors: Ma. Gracia Corazon Cayanan, Kai Yuen Cheong, Li Sha

Abstract:

Training a language model for a minority language has been a challenging task. The lack of available corpora to train and fine-tune state-of-the-art language models is still a challenge in the area of Natural Language Processing (NLP). Moreover, the need for high computational resources and bulk data limit the attainment of this task. In this paper, we presented the following contributions: (1) we introduce and used a translation pair set of Tagalog and English (TL-EN) in pre-training a language model to a minority language resource; (2) we fine-tuned and evaluated top-ranking and pre-trained semantic textual similarity binary task (STSB) models, to both TL-EN and STS dataset pairs. (3) then, we reduced the size of the model to offset the need for high computational resources. Based on our results, the models that were pre-trained to translation pairs and STS pairs can perform well for STSB task. Also, having it reduced to a smaller dimension has no negative effect on the performance but rather has a notable increase on the similarity scores. Moreover, models that were pre-trained to a similar dataset have a tremendous effect on the model’s performance scores.

Keywords: semantic matching, semantic textual similarity binary task, low resource minority language, fine-tuning, dimension reduction, transformer models

Procedia PDF Downloads 211

Authors: Oscar E. Cariceo

Abstract:

In Chile, there is a lack of evidence about the impact of polyvictimization on the emergence of suicidal thoughts among children and young people. Thus, this study aims to explore the association between the episodes of polyvictimization suffered by Chilean children and young people and the manifestation of signs related to suicidal tendencies. To achieve this purpose, secondary data from the First Polyvictimization Survey on Children and Adolescents of 2017 were analyzed, and a binomial logistic regression model was applied to establish the probability that young people are experiencing suicidal ideation episodes. The main findings show that women between the ages of 13 and 15 years, who are in seventh grade and second in subsidized schools, are more likely to express suicidal ideas, which increases if they have suffered different types of victimization, particularly physical violence, psychological aggression, and sexual abuse.

Keywords: Chile, polyvictimization, suicidal ideation, youth

Procedia PDF Downloads 178

Authors: Luh Eka Suryani, Purhadi

Abstract:

Poisson regression is a non-linear regression model with response variable in the form of count data that follows Poisson distribution. Modeling for a pair of count data that show high correlation can be analyzed by Poisson Bivariate Regression. Data, the number of infant mortality and maternal mortality, are count data that can be analyzed by Poisson Bivariate Regression. The Poisson regression assumption is an equidispersion where the mean and variance values are equal. However, the actual count data has a variance value which can be greater or less than the mean value (overdispersion and underdispersion). Violations of this assumption can be overcome by applying Generalized Poisson Regression. Characteristics of each regency can affect the number of cases occurred. This issue can be overcome by spatial analysis called geographically weighted regression. This study analyzes the number of infant mortality and maternal mortality based on conditions in East Java in 2016 using Geographically Weighted Bivariate Generalized Poisson Regression (GWBGPR) method. Modeling is done with adaptive bisquare Kernel weighting which produces 3 regency groups based on infant mortality rate and 5 regency groups based on maternal mortality rate. Variables that significantly influence the number of infant and maternal mortality are the percentages of pregnant women visit health workers at least 4 times during pregnancy, pregnant women get Fe3 tablets, obstetric complication handled, clean household and healthy behavior, and married women with the first marriage age under 18 years.

Keywords: adaptive bisquare kernel, GWBGPR, infant mortality, maternal mortality, overdispersion

Procedia PDF Downloads 160

Authors: Godswill Nwabuisi Osuafor, Sonto Maria Maputle

Abstract:

Background: Abounding studies found that religiosity and social factors modulate alcohol use among university students. However, there is a scarcity of empirical studies examining the protective effects of religiosity and other social factors on alcohol use and abuse in South African universities. The aim of this study was therefore to assess the protective effects of religiosity and roles of social factors on alcohol use among university students. Methodology: A survey on the use of alcohol among 416 university students was conducted using structured questionnaire in 2014. Data were sourced on religiosity and contextual variables. Students were classified as practicing intrinsic religiosity or extrinsic religiosity based on the response to the measures of religiosity. Descriptive, chi square and binary logistic analyses were used in processing the data. Result: Results revealed that alcohol use was associated with religiosity, religion, sex, family history of alcohol use and experimenting with alcohol. Reporting alcohol abuse was significantly predicted by sex, family history of alcohol use and experimenting with alcohol. Religiosity mediated lower alcohol use whereas family history of alcohol use and experimenting with alcohol promoted alcohol use and abuse. Conclusion: Families, religious groups and societal factors may be the specific niches for intervention on alcohol use among university students.

Keywords: religiosity, alcohol use, protective factors, university students

Procedia PDF Downloads 397

Authors: Nicola Rubino

Abstract:

This paper presents an analysis of a group of commodity exporting countries' nominal exchange rate movements in relationship to the US dollar. Using a series of Unrestricted Self-exciting Threshold Autoregressive models (SETAR), we model and evaluate sixteen national CPI price differentials relative to the US dollar CPI. Out-of-sample forecast accuracy is evaluated through calculation of mean absolute error measures on the basis of two-hundred and fifty-three months rolling window forecasts and extended to three additional models, namely a logistic smooth transition regression (LSTAR), an additive non linear autoregressive model (AAR) and a simple linear Neural Network model (NNET). Our preliminary results confirm presence of some form of TAR non linearity in the majority of the countries analyzed, with a relatively higher goodness of fit, with respect to the linear AR(1) benchmark, in five countries out of sixteen considered. Although no model appears to statistically prevail over the other, our final out-of-sample forecast exercise shows that SETAR models tend to have quite poor relative forecasting performance, especially when compared to alternative non-linear specifications. Finally, by analyzing the implied half-lives of the > coefficients, our results confirms the presence, in the spirit of arbitrage band adjustment, of band convergence with an inner unit root behaviour in five of the sixteen countries analyzed.

Keywords: transition regression model, real exchange rate, nonlinearities, price differentials, PPP, commodity points

Procedia PDF Downloads 278

Authors: Leonard Simoni, Ilir Alimehmeti, Ervina Shirka, Endri Hasimi, Ndricim Kallashi, Verona Beka, Suerta Kabili, Artan Goda

Abstract:

Background: Diabetes Mellitus type 2, small vessel calibre, stented length of vessel, complex lesion morphology, and prior bypass surgery have resulted risk factors for In-Stent Restenosis (ISR). However, there are some contradictory results about body mass index (BMI) as a risk factor for ISR. Purpose: We want to identify clinical, lesional and procedural factors that can predict clinical ISR in our patients. Methods: Were enrolled 759 patients who underwent first-time elective PCI with Bare Metal Stents (BMS) from September 2011 to December 2013 in our Department of Cardiology and followed them for at least 1.5 years with a median of 862 days (2 years and 4 months). Only the patients re-admitted with ischemic heart disease underwent control coronary angiography but no routine angiographic control was performed. Patients were categorized in ISR and non-ISR groups and compared between them. Multivariate analysis - Binary Logistic Regression: Forward Conditional Method was used to identify independent predictive risk factors. P was considered statistically significant when <0.05. Results: ISR compared to non-ISR individuals had a significantly lower BMI (25.7±3.3 vs. 26.9±3.7, p=0.004), higher risk anatomy (LM + 3-vessel CAD) (23% vs. 14%, p=0.03), higher number of stents/person used (2.1±1.1 vs. 1.75±0.96, p=0.004), greater length of stents/person used (39.3±21.6 vs. 33.3±18.5, p=0.01), and a lower use of clopidogrel and ASA (together) (95% vs. 99%, p=0.012). They also had a higher, although not statistically significant, prevalence of Diabetes Mellitus (42% vs. 32%, p=0.072) and a greater number of treated vessels (1.36±0.5 vs. 1.26±0.5, p=0.08). In the multivariate analysis, Diabetes Mellitus type 2 and multiple stents used were independent predictors risk factors for In-Stent Restenosis, OR 1.66 [1.03-2.68], p=0.039, and OR 1.44 [1.16-1.78,] p=0.001, respectively. On the other side higher BMI and use of clopidogrel and ASA together resulted protective factors OR 0.88 [0.81-0.95], p=0.001 and OR 0.2 [0.06-0.72] p=0.013, respectively. Conclusion: Diabetes Mellitus and multiple stents are strong predictive risk factors, whereas the use of clopidogrel and ASA together are protective factors for clinical In-Stent Restenosis. Paradoxically High BMI is a protective factor for In-stent Restenosis, probably related to a larger diameter of vessels and consequently a larger diameter of stents implanted in these patients. Further studies are needed to clarify this finding.

Keywords: body mass index, diabetes mellitus, in-stent restenosis, percutaneous coronary intervention

Procedia PDF Downloads 210

Authors: Dana Cassandra Winkler, Delia Leiding, Rene Bergs, Franziska Kaiser, Ramona Kirchhart, Ute Habel

Abstract:

Background: Violence is a multidimensional phenomenon, affecting people of every age, socio-economic status and gender. Nevertheless, most studies primarily focus on men perpetrating women. Aim of the present study is to identify the likelihood of mental and somatic disorders and risk-taking behavior in male violence affected. In addition, the relationship between age of violence experience and the risk for health-related problems was analyzed. Method: On the basis of current evidence, a questionnaire was developed focusing on demographic background, health status, risk-taking behavior, and active and passive violence exposure. In total, 5221 males (Mean: 56,1 years, SD: 17,6) were consulted. To account for the time of violence experience in an efficient way, age clusters ‘0-12 years’, ‘13-20 years’, ‘21-35 years’, ‘36-65 years’ and ‘over 65 years’ were defined. A binary logistic regression was calculated to reveal differences in violence-affected and non-violence affected males regarding health and risk-taking factors. Males who experienced violence on a daily/ almost daily basis vs. males who reported violence occurrence once/ several times a month/ year were compared with respect to health factors and risk-taking behavior. Data of males, who indicated active and passive violence exposure, were analyzed by a chi²-analysis, to investigate a possible relation between the age of victimization and violence perpetration. Findings: Results imply that general violence experience, independent of active and passive violence exposure increases the likelihood in favor of somatic-, psychosomatic- and mental disorders as well as risk-taking behavior in males. Experiencing violence on a daily or almost daily basis in childhood and adolescence may serve as a predictor for increased health problems and risk-taking behavior. Furthermore, the violence experience and perpetration occur significantly within the same age cluster. This underlines the importance of a near-term intervention to minimize the risk, that victims become perpetrators later. Conclusion: The present study reveals predictors concerning health risk factors as well as risk-taking behavior in males with violence exposure. The results of this study may underscore the benefit of intervention and regular health care approaches in violence-affected males and underline the importance of acknowledging the overlap of violence experience and perpetration for further research.

Keywords: health disease, male, mental health, prevalence, risk-taking behavior, violence

Procedia PDF Downloads 212

Authors: Endalamaw Ewnu Kassa, Ade Kurniawan, Ya-Fen Wu, Sajal Biring

Abstract:

Perovskite-based gas sensors represent a highly promising materials within the realm of gas sensing technology, with a particular focus on detecting ammonia (NH3) due to its potential hazards. Our work conducted thorough comparison of various solvents, including dimethylformamide (DMF), DMF-dimethyl sulfoxide (DMSO), dimethylacetamide (DMAC), and DMAC-DMSO, for the preparation of our perovskite solution (MAPbI3). Significantly, we achieved an exceptional response at 10 ppm of ammonia gas by employing a binary solvent mixture of DMAC-DMSO. In contrast to prior reports that relied on single solvents for MAPbI3 precursor preparation, our approach using mixed solvents demonstrated a marked improvement in gas sensing performance. We attained enhanced surface coverage, a reduction in pinhole occurrences, and precise control over grain size in our perovskite films through the careful selection and mixtures of appropriate solvents. This study shows a promising potential of employing binary and multi-solvent mixture strategies as a means to propel advancements in gas sensor technology, opening up new opportunities for practical applications in environmental monitoring and industrial safety.

Keywords: sensors, binary solvents, ammonia, sensitivity, grain size, pinholes, surface coverage

Procedia PDF Downloads 107

Authors: Fidelia A. Orji, Julita Vassileva

Abstract:

This research aims to develop machine learning models for students' academic performance and study strategy prediction, which could be generalized to all courses in higher education. Key learning attributes (intrinsic, extrinsic, autonomy, relatedness, competence, and self-esteem) used in building the models are chosen based on prior studies, which revealed that the attributes are essential in students’ learning process. Previous studies revealed the individual effects of each of these attributes on students’ learning progress. However, few studies have investigated the combined effect of the attributes in predicting student study strategy and academic performance to reduce the dropout rate. To bridge this gap, we used Scikit-learn in python to build five machine learning models (Decision Tree, K-Nearest Neighbour, Random Forest, Linear/Logistic Regression, and Support Vector Machine) for both regression and classification tasks to perform our analysis. The models were trained, evaluated, and tested for accuracy using 924 university dentistry students' data collected by Chilean authors through quantitative research design. A comparative analysis of the models revealed that the tree-based models such as the random forest (with prediction accuracy of 94.9%) and decision tree show the best results compared to the linear, support vector, and k-nearest neighbours. The models built in this research can be used in predicting student performance and study strategy so that appropriate interventions could be implemented to improve student learning progress. Thus, incorporating strategies that could improve diverse student learning attributes in the design of online educational systems may increase the likelihood of students continuing with their learning tasks as required. Moreover, the results show that the attributes could be modelled together and used to adapt/personalize the learning process.

Keywords: classification models, learning strategy, predictive modeling, regression models, student academic performance, student motivation, supervised machine learning

Procedia PDF Downloads 128

Authors: Asrat Agalu Abejew

Abstract:

Background: Antimicrobial resistance (AMR) is a silent tsunami and one of the top global threats to health care and public health. It is one of the common agendas globally and in Ethiopia. Emerging AMR will be a double burden to Ethiopia, which is facing a series of problems from infectious disease morbidity and mortality. In Ethiopia, although there are attempts to document AMR in healthcare institutions, comprehensive and all-inclusive analysis is still lacking. Thus, this study is aimed to determine trends in AMR from 2016-2021. Methods: A retrospective analysis of secondary data recorded in the Amhara Public Health Institute (APHI) from 2016 to 2021 G.C was conducted. Blood, Urine, Stool, Swabs, Discharge, body effusions, and other Microbiological specimens were collected from each study participants, and Bacteria identification and Resistance tests were done using the standard microbiologic procedure. Data was extracted from excel in August 2022, Trends in AMR were analyzed, and the results were described. In addition, the chi-square (X2) test and binary logistic regression were used, and a P. value < 0.05 was used to determine a significant association. Results: During 6 years period, there were 25143 culture and susceptibility tests. Overall, 265 (46.2%) bacteria were resistant to 2-4 antibiotics, 253 (44.2%) to 5-7 antibiotics, and 56 (9.7%) to >=8 antibiotics. The gram-negative bacteria were 166 (43.9%), 155 (41.5%), and 55 (14.6%) resistant to 2-4, 5-7, and ≥8 antibiotics, respectively, whereas 99(50.8%), 96(49.2% and 1 (0.5%) of gram-positive bacteria were resistant to 2-4, 5-7 and ≥8 antibiotics respectively. K. pneumonia 3783 (15.67%) and E. coli 3199 (13.25%) were the most commonly isolated bacteria, and the overall prevalence of AMR was 2605 (59.9%), where K. pneumonia 743 (80.24%), E. cloacae 196 (74.81%), A. baumannii 213 (66.56%) being the most common resistant bacteria for antibiotics tested. Except for a slight decline during 2020 (6469 (25.4%)), the overall trend of AMR is rising from year to year, with a peak in 2019 (8480 (33.7%)) and in 2021 (7508 (29.9%). If left un-intervened, the trend in AMR will increase by 78% of variation from the study period, as explained by the differences in years (R2=0.7799). Ampicillin, Augmentin, ciprofloxacin, cotrimoxazole, tetracycline, and Tobramycin were almost resistant to common bacteria they were tested. Conclusion: AMR is linearly increasing during the last 6 years. If left as it is without appropriate intervention after 15 years (2030 E.C), AMR will increase by 338.7%. A growing number of multi-drug resistant bacteria is an alarm to awake policymakers and those who do have the concern to intervene before it is too late. This calls for a periodic, integrated, and continuous system to determine the prevalence of AMR in commonly used antibiotics.

Keywords: AMR, trend, pattern, MDR

Procedia PDF Downloads 76