Logistic Discrimination Research Articles

Imputation methods on retrospective breast cancer data in Tanzania: A comparative study

Background: Clinical datasets are at risk of having missing data for several reasons including patients’ failure to attend clinical measurements and measurement recorder’s defects. Missing data can significantly affect the analysis and results might be doubtful due to bias caused by omission incomplete records during analysis especially if a dataset is small. This study aims to compare several imputation methods in terms of efficiency in filling-in missing data so as to increase prediction and classification accuracy in breast cancer dataset. Methodology: Five imputation methods namely series mean, k-nearest neighbour, hot deck, predictive mean matching, expected maximisation via bootstrapping, and multiple imputation by chained equations were applied to replace the missing values to the real breast cancer dataset. The efficiency of imputation methods was compared by using the Root Mean Square Errors and Mean Absolute Errors to obtain a suitable complete dataset. Binary logistic regression and linear discrimination classifiers were applied to the imputed dataset to compare their efficacy on classification and discrimination. Results: The evaluation of imputation methods revealed that the predictive mean matching method was better off compared to other imputation methods. In addition, the binary logistic regression and linear discriminant analyses yield almost similar values on overall classification rates, sensitivity and specificity. Conclusion: The predictive mean matching imputation showed higher accuracy in estimating and replacing missing data values in a real breast cancer dataset under the study. It is a more effective and good approach to handle missing data. We recommend replacing missing data by using predictive mean matching since it is a plausible approach toward multiple imputations for numerical variables. It improves estimation and prediction accuracy over the use complete-case analysis especially when percentage of missing data is not very small.

O42 Evaluation of risk prediction models for postoperative pulmonary complications in adults undergoing major abdominal surgery: a systematic review and external validation study in the respiratory complications after abdominal surgery (RECON) cohort

Abstract Introduction Postoperative pulmonary complications (PPCs) following major abdominal surgery result in substantial morbidity and mortality, yet stratifying patients for risk-modifying interventions remains challenging. This study aimed to systematically review and externally validate existing PPC risk prediction models in an international, prospective cohort. Method A systematic search of the MEDLINE and EMBASE databases according to PRISMA guidelines was performed to identify original risk prediction models for PPC following abdominal surgery. Subsequent external validation was performed based on a prospective dataset (REspiratory COmplications after abdomiNal surgery) which encompassed adult patients undergoing major abdominal surgery from January to April 2019 across the UK, Ireland and Australia. The primary outcome was 30-day PPC (StEP-COMPAC criteria definition), and multivariable logistic regression model discrimination were compared (with an area under the curve (AUC) ≥0.7 considered “good”). Result Thirty original risk prediction models were identified from 2819 records, with notable heterogeneity in risk factors considered. Within the validation dataset, the 30-day PPC rate was 7.8% (n = 903/11591), and 6 scores had all variables represented to enable external validation. No score demonstrated statistically significant “good” discrimination for identifying pulmonary complications, with no significant differences between scores. However, the Assess Respiratory Risk in Surgical Patients in Catalonia (ARISCAT) score achieved the highest discrimination (AUC = 0.709, 95% CI = 0.692–0.727). Conclusion The risk of PPC for patients following major abdominal surgery in the pre-covid era is not well described by existing prediction tools. New prediction tools are required to account for additional variation introduced for patients affected by SARS-CoV-2 infection. Take-home Message The risk of PPC for patients following major abdominal surgery in the pre-covid era is not well described by existing prediction tools. New prediction tools are required to account for additional variation introduced for patients affected by SARS-CoV-2 infection.

Softmax Model as Generalization upon Logistic Discrimination Suffers from Overfitting

The motivation behind this paper is to investigate the use of Softmax model for classification. We show that Softmax model is a nonlinear generalization for the logistic discrimination, that can approximate the posterior probabilities of classes where other Artificial neural network (ANN) models don't have this ability. We show that Softmax model has more flexibility than logistic discrimination in terms of correct classification. To show the performance of Softmax model a medical data set on thyroid gland state is used. The result is that Softmax model may suffer from overfitting.

The Varying Threshold Values of Logistic Regression and Linear Discriminant for Classifying Fraudulent Firm

The aim of the research is to find the best performance both of logistic regression and linear discriminant which their threshold uses some various values. The performance tools used for evaluating classifier model are confusion matrix, precision-recall, F1 score and receiver operation characteristic (ROC) curve. The Audit-risk data set are used for the implementation of the proposed method. The screening data and dimension reduction by using principal component analysis (PCA) are the first step that must be conducted before the data are divided into the training and testing set. After the training process for obtaining the classifier model parameters has been completed, the calculation of performance measures is done only on the testing set where the various constants are added to the threshold value of both classifier models. The logistic regression classifier has the best performance of 94% on the precision-recall, 91.7% on the F1-score, and 0.906 on the area under curve (AUC) where the threshold values are on the interval between 0.002 and 0.018. On the other hand, the linear discriminant classifier has the best performance when the threshold value is 0.035 and its performance value is respectively the precision-recall of 94%, the F1-score of 91.7%, and the AUC of 0.846.

Using logistics models to study the most important factors affecting diabetes by type of disease

In this research, analysis of the discrimination of logistics is used, and the maximum likelihood method was used to estimate the parameters of the logistic model. This study has been done on diabetic patents visiting laila kassim diabetes center in hawler city in 2018. The data has been collected from both types of diabetes (type I and type II) via filing on questionnaire during personal interview with every patients and historical records. The main objective of this research is to study the main risk factors that differentiate between type I and type II and through using statistical analysis of the discrimination of logistics we can discover the effect of either factor on increasing risk (incidence) of diabetic. We concluded in this study, the model of logistic discrimination is appropriate in the study of the factors affecting for both types of diabetes and found that the estimated parameters of the logistic model are significant through the test (Wald). The two variables (the age of incidence of diabetes and blood sugar level) are significant factors on dependable variable which represent the incidence of diabetes while the other variables have no effect. And the correct classification ratio is )83.6%(.

THEFT DETECTION IN POWER UTILITIES USING ENSEMBLE OF CHAID DECISION TREE ALGORITHM

Theft of electricity and fraud in energy consumption billing are the primary concerns for Distribution System Operators . Because of those illegal activities, it is believed that billions of dollars are wasted each year. DSOs around the world continue to use conventional time consuming and inefficient methods for non-technical loss detection, particularly in underdeveloped countries . This research work attempts to solve the problems as mentioned above by designing an effective model for detecting electricity theft to classify fraudster customers in a power delivery system. The key motivation for this current study is to support the DSOs in their fight against the stealing of electricity. Initially, the proposed method uses the monthly energy customer consumption data obtained from Multan Electric Power Company (MEPCO) Pakistan to segregate fraudsters and honest customers. The Bagged Chi - square Automatic Interaction Detection (CHAID) based decision tree algorithm is then used to classify the honest and fraudster consumers.Furthermore, based on the mentioned metrics, the performance superiority of the Bagged CHAID-based NTL detection method is validated by comparing its efficacy with that of few well-known state-of-the-art machine learning algorithms such as Artificial Neural Network (ANN), Support Vector Machine (SVM),) Logistic Regression (LR), Discriminant Analysis and Bayesian Network (BN).

Mathematical modeling for the prediction of cerebral white matter lesions based on clinical examination data.

Cerebral white matter lesions are ischemic symptoms caused mainly by microangiopathy; they are diagnosed by MRI because they show up as abnormalities in MRI images. Because patients with white matter lesions do not have any symptoms, MRI often detects the lesions for the first time. Generally, head MRI for the diagnosis and grading of cerebral white matter lesions is performed as an option during medical checkups in Japan. In this study, we develop a mathematical model for the prediction of white matter lesions using data from routine medical evaluations that do not include a head MRI. Linear discriminant analysis, logistic discrimination, Naive Bayes classifier, support vector machine, and random forest were investigated and evaluated by ten-fold cross-validation, using clinical data for 1,904 examinees (988 males and 916 females) from medical checkups that did include the head MRI. The logistic regression model was selected based on a comparison of accuracy and interpretability. The model variables consisted of age, gender, plaque score (PS), LDL, systolic blood pressure (SBP), and administration of antihypertensive medication (odds ratios: 2.99, 1.57, 1.18, 1.06, 1.12, and 1.52, respectively) and showed Areas Under the ROC Curve (AUC) 0.805, the model displayed sensitivity of 72.0%, and specificity 75.1% when the most appropriate cutoff value was used, 0.579 as given by the Youden Index. This model has shown to be useful to identify patients with a high-risk of cerebral white matter lesions, who can then be diagnosed with a head MRI examination in order to prevent dementia, cerebral infarction, and stroke.

Functional logistic discrimination with sparse PCA and its application to the structural MRI

We propose a functional classification method with high-dimensional image predictors using a combination of logistic discrimination and basis expansions with sparse principal component analysis (PCA). Our model is an extension of the existing functional generalized linear models with image predictors using functional principal component regression to L1-regularized principal components. This extension enables us to create a more flexible prognostic region that does not depend on the shape of basis functions. Monte Carlo simulations were conducted to examine the method’s efficiency when compared with several possible classification techniques. Our method was shown to be the best in terms of both sensitivity and specificity for detecting the shape of interests and classifying groups. In addition, our model was applied to data on Alzheimer’s disease. Our model detected the prognostic brain region and was used to classify early-stage Alzheimer patients efficiently, based on three-dimensional structural magnetic resonance imaging (sMRI).

Tipología influyente en el rendimiento académico de alumnos universitarios

El presente trabajo aborda un informe estadístico centrado en caracterizar el rendimiento académico de alumnos universitarios, a partir de la determinación de variables asociadas, aplicando técnicas estadísticas del Análisis Multivariado. Los análisis efectuados se basan en datos provenientes de una encuesta realizada en el año 2015, a los alumnos de la Facultad de Ciencias Exactas Físicas y Naturales, de la Facultad de Filosofía, Humanidades y Artes de la Universidad Nacional de San Juan, Argentina. Mediante un Análisis Factorial de Correspondencias Múltiples, Análisis de Conglomerados y Análisis de Discriminación Logística, se pudieron identificar tipologías de alumnos y variables influyentes que diferencian a los alumnos según su rendimiento. Los resultados aportan herramientas que permiten realizar un válido diagnóstico para orientar de manera efectiva las intervenciones que realice la institución educativa.

Classifying next-generation sequencing data using a zero-inflated Poisson model.

With the development of high-throughput techniques, RNA-sequencing (RNA-seq) is becoming increasingly popular as an alternative for gene expression analysis, such as RNAs profiling and classification. Identifying which type of diseases a new patient belongs to with RNA-seq data has been recognized as a vital problem in medical research. As RNA-seq data are discrete, statistical methods developed for classifying microarray data cannot be readily applied for RNA-seq data classification. Witten proposed a Poisson linear discriminant analysis (PLDA) to classify the RNA-seq data in 2011. Note, however, that the count datasets are frequently characterized by excess zeros in real RNA-seq or microRNA sequence data (i.e. when the sequence depth is not enough or small RNAs with the length of 18-30 nucleotides). Therefore, it is desired to develop a new model to analyze RNA-seq data with an excess of zeros. In this paper, we propose a Zero-Inflated Poisson Logistic Discriminant Analysis (ZIPLDA) for RNA-seq data with an excess of zeros. The new method assumes that the data are from a mixture of two distributions: one is a point mass at zero, and the other follows a Poisson distribution. We then consider a logistic relation between the probability of observing zeros and the mean of the genes and the sequencing depth in the model. Simulation studies show that the proposed method performs better than, or at least as well as, the existing methods in a wide range of settings. Two real datasets including a breast cancer RNA-seq dataset and a microRNA-seq dataset are also analyzed, and they coincide with the simulation results that our proposed method outperforms the existing competitors. The software is available at http://www.math.hkbu.edu.hk/∼tongt. xwan@comp.hkbu.edu.hk or tongt@hkbu.edu.hk. Supplementary data are available at Bioinformatics online.

Specific plasma amino acid disturbances associated with metabolic syndrome

PurposeThe primary objective of the present study was to examine the association between branched chain and aromatic amino acid profiles (BCAA and AAA respectively) and the metabolic syndrome (MS), and to evaluate the clinical utility of these associations in the diagnostic process.MethodsTwo hundred and sixty three healthy men with MS [MS(+): n = 165] and without MS [MS(−): n = 98] were enrolled in the observational study. Anthropometrical, biochemical, and amino acid measurements were performed. The ability of the BCAA and AAA to discriminate subjects with MS and insulin resistance was tested. Based on logistic discrimination, a multivariate early MS diagnostic model was built, and its discrimination properties were evaluated.ResultsTwo functionally independent amino acid clusters were identified. BCAA and phenylalanine differed significantly between MS(+) and MS(−) participants (P = 0.003). These factors were also found to be indicators of MS(+) individuals (AUC: 0.66; 95% CI: 0.5757–0.7469), and correlated with cardiometabolic factors. No statistically significant differences in amino acid concentrations between those with and without insulin resistance were noted, and none of the amino groups were indicators of insulin resistance. The proposed MS multivariate diagnostic model consisted of phenylalanine, insulin, leptin, and adiponectin, and had good discrimination properties [AUC 0.79; 95% CI: 0.7239–0.8646].ConclusionsMS is associated with selective BCAA and AAA profile disturbances, which could be part of cardiometabolic disease pathogenesis and derive neither directly from insulin sensitivity impairment, nor obesity or muscle mass. The MS diagnostic model developed and described herein should be validated in future studies.

Development of a new risk nomogram of perioperative major adverse cardiac events for Chinese patients undergoing colorectal carcinoma surgery.

The purpose of this study is to create a new risk nomogram to predict perioperative major adverse cardiac events in patients undergoing colorectal carcinoma surgery. A total of 1899 patients who underwent colorectal carcinoma surgery at a tertiary teaching hospital in China between 2007 and 2012 were recruited. Logistic regression analysis was used to define risk factors for major adverse cardiac events. A nomogram-predicting model was built based on the logistic regression model and discrimination was tested by receiver operating characteristic curves. Fifty-six (2.9%) among 1899 included patients developed at least one cardiac event. Eight risk factors were found in the multivariate logistic regression model, which included age ≥60years, smoking, a history of chronic kidney disease, coronary artery disease, congestive heart failure, hypertension, preoperative albumin levels ≤35g/L, blood transfusion ≥500mL, and intraoperative blood pressure variability. P=0.708 in the Hosmer-Lemeshow test indicated acceptable calibration power. Based on this multivariate model, we built a risk nomogram model for these cardiac events with an area under the curve (95% confidence interval) of 0.923 (0.889, 0.957), which demonstrated good discrimination of this model. When the probability cutoff was 1.9% (total score of 83), the nomogram model had the best sensitivity and specificity in predicting cardiac events. A new nomogram model for predicting perioperative major adverse cardiac events in patients who had colorectal carcinoma surgery was established in this study. When the total score is >83, patients undergoing colorectal carcinoma surgery should be considered at high risk of perioperative major adverse cardiac events.

Asymptotic Performance of the Location and Logistic Classification Rules for Multivariate Binary Variables

This paper focuses on the Asymptotic Classification Procedures in Two Group Discriminate Analysis with Multivariate Binary Variables. Two data patterns were simulated using the R-Software Statistical Analysis System 2.15.3 and was subjected to two linear classification namely; Location and Logistic Models. To judge the performance of these models, the apparent error rates for each procedure are obtained for different sample sizes. The results obtained show that the location model performed better than Logistic Discrimination with the variation in the error rates being higher for Logistic Discrimination rule.

Logistic discrimination based on G-mean and F-measure for imbalanced problem

As a well known statistical method, logistic discrimination has been successfully used in many practical applications including medical diagnosis and personal credit assessment. In this paper, we apply this model to imbalanced problem which is also referred to as skewed or rare class problem, characterized by having many more instances of one class (negative class or majority class) than the other (positive class or minority class). However, traditional logistic discrimination tries to pursue a high accuracy by assuming that all classes have similar size, leading to the fact that instances with positive classes are often overlooked and misclassified to negative ones. To fully consider class imbalance, we re-learn the two basic measures for imbalanced problem, g-mean and f-measure, and design two new cost functions, i.e., g-mean based metric (GM) and f-measure based metric (FM), to supervise logistic discrimination learning the corresponding parameters, where GM is the geometric mean estimation of recall of both positive and negative class as g-mean and FM is a harmonic mean between recall and precision of positive class as f-measure. The experiments on UCI data sets show that the proposed method presents significant advantage comparing to state-of-the-art classification methods on all metrics used in this paper including accuracy, recall, f-measure and g-mean.

Development of a real time activity monitoring Android application utilizing SmartStep.

Footwear based activity monitoring systems are becoming popular in academic research as well as consumer industry segments. In our previous work, we had presented developmental aspects of an insole based activity and gait monitoring system-SmartStep, which is a socially acceptable, fully wireless and versatile insole. The present work describes the development of an Android application that captures the SmartStep data wirelessly over Bluetooth Low energy (BLE), computes features on the received data, runs activity classification algorithms and provides real time feedback. The development of activity classification methods was based on the the data from a human study involving 4 participants. Participants were asked to perform activities of sitting, standing, walking, and cycling while they wore SmartStep insole system. Multinomial Logistic Discrimination (MLD) was utilized in the development of machine learning model for activity prediction. The resulting classification model was implemented in an Android Smartphone. The Android application was benchmarked for power consumption and CPU loading. Leave one out cross validation resulted in average accuracy of 96.9% during model training phase. The Android application for real time activity classification was tested on a human subject wearing SmartStep resulting in testing accuracy of 95.4%.

Imbalanced Learning Based on Logistic Discrimination.

In recent years, imbalanced learning problem has attracted more and more attentions from both academia and industry, and the problem is concerned with the performance of learning algorithms in the presence of data with severe class distribution skews. In this paper, we apply the well-known statistical model logistic discrimination to this problem and propose a novel method to improve its performance. To fully consider the class imbalance, we design a new cost function which takes into account the accuracies of both positive class and negative class as well as the precision of positive class. Unlike traditional logistic discrimination, the proposed method learns its parameters by maximizing the proposed cost function. Experimental results show that, compared with other state-of-the-art methods, the proposed one shows significantly better performance on measures of recall, g-mean, f-measure, AUC, and accuracy.

Estimating a Logistic Discrimination Functions When One of the Training Samples Is Subject to Misclassification: A Maximum Likelihood Approach.

The problem of discrimination and classification is central to much of epidemiology. Here we consider the estimation of a logistic regression/discrimination function from training samples, when one of the training samples is subject to misclassification or mislabeling, e.g. diseased individuals are incorrectly classified/labeled as healthy controls. We show that this leads to zero-inflated binomial model with a defective logistic regression or discrimination function, whose parameters can be estimated using standard statistical methods such as maximum likelihood. These parameters can be used to estimate the probability of true group membership among those, possibly erroneously, classified as controls. Two examples are analyzed and discussed. A simulation study explores properties of the maximum likelihood parameter estimates and the estimates of the number of mislabeled observations.

Effective moving shadow detection using statistical discriminant model

Shadow detection is considered to be a challenging problem in computer vision and video analysis recently. In this paper, a novel moving shadow detection method based on statistical discriminant model for surveillance scenarios is presented. First, we extract various features from video frames to form a feature space for classification. Second, we adopt a statistical learning method that employs Partial Least Squares (PLS) and Logistic Discrimination (LD) to classify moving shadows and their corresponding objects. Finally, simple post processing is performed to obtain the refined classified results. Extensive experimental results indicate that the proposed method can adapt to different kinds of scenarios under various illumination conditions automatically. Specifically, compared with several well-known methods, our method exhibits much better performance for moving shadow detection.

Modeling and mapping the probability of occurrence of invasive wild pigs across the contiguous United States.

Wild pigs (Sus scrofa), also known as wild swine, feral pigs, or feral hogs, are one of the most widespread and successful invasive species around the world. Wild pigs have been linked to extensive and costly agricultural damage and present a serious threat to plant and animal communities due to their rooting behavior and omnivorous diet. We modeled the current distribution of wild pigs in the United States to better understand the physiological and ecological factors that may determine their invasive potential and to guide future study and eradication efforts. Using national-scale wild pig occurrence data reported between 1982 and 2012 by wildlife management professionals, we estimated the probability of wild pig occurrence across the United States using a logistic discrimination function and environmental covariates hypothesized to influence the distribution of the species. Our results suggest the distribution of wild pigs in the U.S. was most strongly limited by cold temperatures and availability of water, and that they were most likely to occur where potential home ranges had higher habitat heterogeneity, providing access to multiple key resources including water, forage, and cover. High probability of occurrence was also associated with frequent high temperatures, up to a high threshold. However, this pattern is driven by pigs’ historic distribution in warm climates of the southern U.S. Further study of pigs’ ability to persist in cold northern climates is needed to better understand whether low temperatures actually limit their distribution. Our model highlights areas at risk of invasion as those with habitat conditions similar to those found in pigs’ current range that are also near current populations. This study provides a macro-scale approach to generalist species distribution modeling that is applicable to other generalist and invasive species.

Posture and activity recognition and energy expenditure estimation in a wearable platform.

The use of wearable sensors coupled with the processing power of mobile phones may be an attractive way to provide real-time feedback about physical activity and energy expenditure (EE). Here, we describe the use of a shoe-based wearable sensor system (SmartShoe) with a mobile phone for real-time recognition of various postures/physical activities and the resulting EE. To deal with processing power and memory limitations of the phone, we compare the use of support vector machines (SVM), multinomial logistic discrimination (MLD), and multilayer perceptrons (MLP) for posture and activity classification followed by activity-branched EE estimation. The algorithms were validated using data from 15 subjects who performed up to 15 different activities of daily living during a 4-h stay in a room calorimeter. MLD and MLP demonstrated activity classification accuracy virtually identical to SVM (∼ 95%) while reducing the running time and the memory requirements by a factor of >10 3. Comparison of per-minute EE estimation using activity-branched models resulted in accurate EE prediction (RMSE = 0.78 kcal/min for SVM and MLD activity classification, 0.77 kcal/min for MLP versus RMSE of 0.75 kcal/min for manual annotation). These results suggest that low-power computational algorithms can be successfully used for real-time physical activity monitoring and EE estimation on a wearable platform.