Prognostic modelling with logistic regression analysis: a comparison of selection and estimation methods in small data sets

2000 ◽  
Vol 19 (8) ◽  
pp. 1059-1079 ◽  
Author(s):  
Ewout W. Steyerberg ◽  
Marinus J.C. Eijkemans ◽  
Frank E. Harrell ◽  
J. Dik F. Habbema
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Estimation Methods ◽  
Small Data ◽  
Data Sets ◽  
Small Data Sets ◽  
Prognostic Modelling

Counting the Hidden Defects in Software Documents

2010 ◽  
pp. 519-538
Author(s):  
Frank Padberg
Keyword(s):  
Neural Networks ◽  
Training Data ◽  
Estimation Methods ◽  
Model Parameters ◽  
Small Data ◽  
Data Sets ◽  
Software Inspection ◽  
Bayesian Techniques ◽  
Standard Quality ◽  
Small Data Sets

The author uses neural networks to estimate how many defects are hidden in a software document. Input for the models are metrics that get collected when effecting a standard quality assurance technique on the document, a software inspection. For inspections, the empirical data sets typically are small. The author identifies two key ingredients for a successful application of neural networks to small data sets: Adapting the size, complexity, and input dimension of the networks to the amount of information available for training; and using Bayesian techniques instead of cross-validation for determining model parameters and selecting the final model. For inspections, the machine learning approach is highly successful and outperforms the previously existing defect estimation methods in software engineering by a factor of 4 in accuracy on the standard benchmark. The author’s approach is well applicable in other contexts that are subject to small training data sets.

Latent Class Modeling with Covariates: Two Improved Three-Step Approaches

2010 ◽  
Vol 18 (4) ◽  
pp. 450-469 ◽  
Author(s):  
Jeroen K. Vermunt
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Latent Class ◽  
Correction Method ◽  
Estimation Methods ◽  
Parameter Estimates ◽  
Sampling Variance ◽  
Class Membership ◽  
Ml Estimation

Researchers using latent class (LC) analysis often proceed using the following three steps: (1) an LC model is built for a set of response variables, (2) subjects are assigned to LCs based on their posterior class membership probabilities, and (3) the association between the assigned class membership and external variables is investigated using simple cross-tabulations or multinomial logistic regression analysis. Bolck, Croon, and Hagenaars (2004) demonstrated that such a three-step approach underestimates the associations between covariates and class membership. They proposed resolving this problem by means of a specific correction method that involves modifying the third step. In this article, I extend the correction method of Bolck, Croon, and Hagenaars by showing that it involves maximizing a weighted log-likelihood function for clustered data. This conceptualization makes it possible to apply the method not only with categorical but also with continuous explanatory variables, to obtain correct tests using complex sampling variance estimation methods, and to implement it in standard software for logistic regression analysis. In addition, a new maximum likelihood (ML)—based correction method is proposed, which is more direct in the sense that it does not require analyzing weighted data. This new three-step ML method can be easily implemented in software for LC analysis. The reported simulation study shows that both correction methods perform very well in the sense that their parameter estimates and their SEs can be trusted, except for situations with very poorly separated classes. The main advantage of the ML method compared with the Bolck, Croon, and Hagenaars approach is that it is much more efficient and almost as efficient as one-step ML estimation.

PREVALENCE OF CHILD MARRIAGE AMONG BANGLADESHI WOMEN AND TREND OF CHANGE OVER TIME

2015 ◽  
Vol 48 (4) ◽  
pp. 530-538 ◽  
Author(s):  
Md. Golam Hossain ◽  
Rashidul Alam Mahumud ◽  
Aik Saw
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Rural Areas ◽  
Economic Status ◽  
Binary Logistic Regression ◽  
Child Marriage ◽  
Binary Logistic Regression Analysis ◽  
Data Sets ◽  
First Marriage

SummaryMany Bangladeshi women marry early, and many marry before the legal age of 18 years. This practice has been associated with a higher risk of health and medical morbidities, and also early pregnancy with higher pre- and postnatal complications. The aim of this study was to determine the prevalence, and factors associated with, child marriage among Bangladeshi women using multiple binary logistic regression analysis of data from the BDHS-2011. Further analysis on the trend of age at first marriage was performed with additional data sets from previous surveys. The mean and median of ages at first marriage of Bangladeshi women in 2011 were 15.69±2.97 and 15.00 years, respectively. A remarkably high percentage (78.2%) married before the age of 18; of these, 5.5% married at a very early age (before 13 years of age). Binary logistic regression analysis demonstrated that uneducated women were more likely to be married early (p<0.001) than those with secondary and higher education. Child marriage was especially pronounced among women with uneducated husbands, Muslims, those with poor economic backgrounds and those living in rural areas. Further analysis including data from previous BDHS surveys showed that child marriage among Bangladeshi women had a decreasing trend from 1993–94 to 2011. These results show that child marriage was very common in Bangladesh, and closely associated with low level of education and low economic status. The decreasing trend in child marriage indicates an improvement over the past two decades but more effort is needed to further reduce and eventually eliminate the practice.

ANALISIS HUBUNGAN SARANA, PENGETAHUAN, DAN SIKAP BIDAN PRAKTIK MANDIRI DENGAN KEPATUHAN MENJALANKAN STANDAR PELAYANAN ANTENATAL CARE DI KOTA PADANGSIDIMPUAN

2019 ◽  
Vol 2 (1) ◽  
pp. 27-33
Author(s):  
Megawati Sinambela ◽  
Evi Erianty Hasibuan
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Antenatal Care ◽  
Pregnant Women ◽  
Logistic Regression Analysis ◽  
Care Service ◽  
Multivariate Logistic Regression Analysis ◽  
Bivariate Analysis ◽  
Cross Sectional ◽  
The North

Antenatal care is a service provided to pregnant women to monitor, support maternal health and detect mothers whether normal or problematic pregnant women. According to the WHO, globally more than 70% of maternal deaths are caused by complications of pregnancy and childbirth such as hemorrhage, hypertension, sepsis, and abortion. Based on data obtained from the profile of the North Sumatra provincial health office in 2017, in the city of Padangsidimpuan in 2017 the coverage of ANC visits reached (76.58%) and had not reached the target in accordance with the 2017 Provincial Health Office strategy plan (95%). This type of research was an observational analytic study with a cross sectional design. The population in this study were independent practice midwives who were in the Padangsidimpuan, the sample in this study amounted to 102 respondents. The technique of collecting data used questionnaires and data analysis used univariate, bivariate and multivariate analysis with logistic regression analysis. Based on bivariate analysis showed that there was a relationship between facilities, knowledge and attitudes of independent midwives with compliance with the standards of antenatal care services with a value of p <0.05. The results of the study with multivariate logistic regression analysis showed that the factors associated with the compliance of independent midwives in carrying out antenatal care service standards were attitudes with values (p = 0.026).

scholarly journals Efficacy of flurbiprofen axetil for preventing postanesthetic shivering in patients undergoing gynecologic laparotomy surgeries

2020 ◽  
Vol 6 (1) ◽  
Author(s):  
Atsushi Kotera
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Postoperative Pain ◽  
Logistic Regression Analysis ◽  
Pain Control ◽  
Multivariate Logistic Regression Analysis ◽  
Multivariate Logistic Regression ◽  
Postoperative Pain Control ◽  
Flurbiprofen Axetil ◽  
Postanesthetic Shivering

Abstract Background Postanesthetic shivering is an unpleasant adverse event in surgical patients. A nonsteroidal anti-inflammatory drug has been reported to be useful in preventing postanesthetic shivering in several previous studies. The aim of this study was to evaluate the efficacy of flurbiprofen axetil being a prodrug of a nonsteroidal anti-inflammatory drug for preventing postanesthetic shivering in patients undergoing gynecologic laparotomy surgeries. Method This study is a retrospective observational study. I collected data from patients undergoing gynecologic laparotomy surgeries performed between October 1, 2019, and September 30, 2020, at Kumamoto City Hospital. All the patients were managed with general anesthesia with or without epidural analgesia. The administration of intravenous 50 mg flurbiprofen axetil for postoperative pain control at the end of the surgery was left to the individual anesthesiologist. The patients were divided into two groups: those who had received intravenous flurbiprofen axetil (flurbiprofen group) and those who had not received intravenous flurbiprofen axetil (non-flurbiprofen group), and I compared the frequency of postanesthetic shivering between the two groups. Additionally, the factors presumably associated with postanesthetic shivering were collected from the medical charts. Intergroup differences were assessed with the χ2 test with Yates’ correlation for continuity category variables. The Student’s t test was used to test for differences in continuous variables. Furthermore, a multivariate logistic regression analysis was performed to elucidate the relationship between the administration of flurbiprofen axetil and the incidence of PAS. Results I retrospectively examined the cases of 141 patients aged 49 ± 13 (range 21-84) years old. The overall postanesthetic shivering rate was 21.3% (30 of the 141 patients). The frequency of postanesthetic shivering in the flurbiprofen group (n = 31) was 6.5%, which was significantly lower than that in the non-flurbiprofen group (n = 110), 25.5% (p value = 0.022). A multivariate logistic regression analysis showed that administration of flurbiprofen axetil was independently associated with a reduced incidence of postanesthetic shivering (odds ratio 0.12; 95% confidence interval, 0.02-0.66, p value = 0.015). Conclusions My result suggests that intraoperative 50 mg flurbiprofen axetil administration for postoperative pain control is useful to prevent postanesthetic shivering in patients undergoing gynecologic laparotomy surgeries.

scholarly journals Prior Admissions as a Risk Factor for Readmission in Patients Surgically Treated for Femur Fractures: Implications for a Potential Hip Fracture Bundle

2021 ◽  
Vol 12 ◽  
pp. 215145932199616
Author(s):  
Robert Erlichman ◽  
Nicholas Kolodychuk ◽  
Joseph N. Gabra ◽  
Harshitha Dudipala ◽  
Brook Maxhimer ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Regression Analysis ◽  
Hip Fracture ◽  
Logistic Regression Analysis ◽  
Hip Fractures ◽  
Surgical Procedure ◽  
Hospital Admissions ◽  
Fracture Care ◽  
Increased Risk ◽  
Payment Models

Introduction: Hip fractures are a significant economic burden to our healthcare system. As there have been efforts made to create an alternative payment model for hip fracture care, it will be imperative to risk-stratify reimbursement for these medically comorbid patients. We hypothesized that patients readmitted to the hospital within 90 days would be more likely to have a recent previous hospital admission, prior to their injury. Patients with a recent prior admission could therefore be considered higher risk for readmission and increased cost. Methods: A retrospective chart review identified 598 patients who underwent surgical fixation of a hip or femur fracture. Data on readmissions within 90 days of surgical procedure and previous admissions in the year prior to injury resulting in surgical procedure were collected. Logistic regression analysis was used to determine if recent prior admission had increased risk of 90-day readmission. A subgroup analysis of geriatric hip fractures and of readmitted patients were also performed. Results: Having a prior admission within one year was significantly associated (p < 0.0001) for 90-day readmission. Specifically, logistic regression analysis revealed that a prior admission was significantly associated with 90-day readmission with an odds ratio of 7.2 (95% CI: 4.8-10.9). Discussion: This patient population has a high rate of prior hospital admissions, and these prior admissions were predictive of 90-day readmission. Alternative payment models that include penalties for readmissions or fail to apply robust risk stratification may unjustly penalize hospital systems which care for more medically complex patients. Conclusions: Hip fracture patients with a recent prior admission to the hospital are at an increased risk for 90-day readmission. This information should be considered as alternative payment models are developed for hip fracture care.

scholarly journals Classification of jujube defects in small data sets based on transfer learning

2021 ◽  
Author(s):  
Jianping Ju ◽  
Hong Zheng ◽  
Xiaohang Xu ◽  
Zhongyuan Guo ◽  
Zhaohui Zheng ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Loss Function ◽  
Training Model ◽  
Parameter Distribution ◽  
Test Accuracy ◽  
Small Data ◽  
Data Sets ◽  
Data Set ◽  
Small Data Sets

AbstractAlthough convolutional neural networks have achieved success in the field of image classification, there are still challenges in the field of agricultural product quality sorting such as machine vision-based jujube defects detection. The performance of jujube defect detection mainly depends on the feature extraction and the classifier used. Due to the diversity of the jujube materials and the variability of the testing environment, the traditional method of manually extracting the features often fails to meet the requirements of practical application. In this paper, a jujube sorting model in small data sets based on convolutional neural network and transfer learning is proposed to meet the actual demand of jujube defects detection. Firstly, the original images collected from the actual jujube sorting production line were pre-processed, and the data were augmented to establish a data set of five categories of jujube defects. The original CNN model is then improved by embedding the SE module and using the triplet loss function and the center loss function to replace the softmax loss function. Finally, the depth pre-training model on the ImageNet image data set was used to conduct training on the jujube defects data set, so that the parameters of the pre-training model could fit the parameter distribution of the jujube defects image, and the parameter distribution was transferred to the jujube defects data set to complete the transfer of the model and realize the detection and classification of the jujube defects. The classification results are visualized by heatmap through the analysis of classification accuracy and confusion matrix compared with the comparison models. The experimental results show that the SE-ResNet50-CL model optimizes the fine-grained classification problem of jujube defect recognition, and the test accuracy reaches 94.15%. The model has good stability and high recognition accuracy in complex environments.

scholarly journals Study on the relationship of depression, anxiety, lifestyle and eating habits with the severity of reflux esophagitis

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Rongxin Wang ◽  
Jing Wang ◽  
Shuiqing Hu
Keyword(s):  
Quality Of Life ◽  
Logistic Regression ◽  
Regression Analysis ◽  
Logistic Regression Analysis ◽  
Reflux Esophagitis ◽  
Eating Habits ◽  
Positive Correlation ◽  
Relationship Of ◽  
The Relationship

Abstract Background The etiology of reflux esophagitis (RE) is multi-factorial. This study analyzed the relationship of depression, anxiety, lifestyle and eating habits with RE and its severity and further explored the impact of anxiety and depression on patients’ symptoms and quality of life. Methods From September 2016 to February 2018, a total of 689 subjects at Xuanwu Hospital Capital Medical University participated in this survey. They were divided into the RE group (patients diagnosed with RE on gastroscopy, n = 361) and the control group (healthy individuals without heartburn, regurgitation and other gastrointestinal symptoms, n = 328). The survey included general demographic information, lifestyle habits, eating habits, comorbidities, current medications, the gastroesophageal reflux disease (GERD) questionnaire (GerdQ), the Patient Health Questionnaire-9 depression scale and the General Anxiety Disorder-7 anxiety scale. Results The mean age and sex ratio of the two groups were similar. Multivariate logistic regression analysis identified the following factors as related to the onset of RE (p < 0.05): low education level; drinking strong tea; preferences for sweets, noodles and acidic foods; sleeping on a low pillow; overeating; a short interval between dinner and sleep; anxiety; depression; constipation; history of hypertension; and use of oral calcium channel blockers. Ordinal logistic regression analysis revealed a positive correlation between sleeping on a low pillow and RE severity (p = 0.025). Depression had a positive correlation with the severity of symptoms (rs = 0.375, p < 0.001) and patients’ quality of life (rs = 0.306, p < 0.001), whereas anxiety showed no such association. Conclusions Many lifestyle factors and eating habits were correlated with the onset of RE. Notably, sleeping on a low pillow was positively correlated with RE severity, and depression was positively related to the severity of symptoms and patients’ quality of life.

Sign in / Sign up

Export Citation Format

Share Document