scholarly journals Comparison of Neural Networks and Logistic Regression in Assessing the Occurrence of Failures in Steel Structures of Transmission Lines

2016 ◽  
Vol 10 (1) ◽  
pp. 11-26
Author(s):  
A.C.G Bissacot ◽  
S.A.B Salgado ◽  
P.P Balestrassi ◽  
A.P Paiva ◽  
A.C Zambroni Souza ◽  
...  
Keyword(s):  
Neural Networks ◽  
Logistic Regression ◽  
Transmission Lines ◽  
Steel Structures ◽  
Roc Curves ◽  
Operating Voltage ◽  
Operating Characteristics ◽  
Metal Structures ◽  
Land Type ◽  
Classic Technique

In this work, we evaluate the probability of falling metal structures from transmission lines. It is our objective to extract knowledge about which variables influence the mechanical behavior of the operating lines and can be used to diagnose potential falling towers. Those pieces of information can become a basis for directing the investments of reinforcement structures, avoiding the occurrence of long turn offs and high costs as a consequence of damage to towers of transmission lines. The results are obtained using the history of 181 metal structures currently in operation in the state of Paraná/Brazil. For the classification of transmission lines susceptible to failures it is proposed to identify the most likely lines considering the following parameters: operating voltage, wind and relief of the region, air masses, temperature, land type, mechanical capacity, function and foundation structure. The classic technique of classifying binary events used in this type of problem is the logistic regression (LR). The more recent technique for classification, using Artificial Neural Networks (ANN) can also be applied. The results are compared through the area under receiver operating characteristics (ROC) curves.

scholarly journals Landslide Susceptibility Mapping Using Analytical Hierarchy Process, Statistical Index, Index of Enthropy, and Logistic Regression Approaches in the TinalahWatershed, Yogyakarta

2018 ◽  
Vol 2 (2) ◽  
pp. 67
Author(s):  
Tee Xiong ◽  
I Gde Budi Indrawan ◽  
Doni Prakasa Eka Putra
Keyword(s):  
Logistic Regression ◽  
Analytical Hierarchy Process ◽  
Landslide Susceptibility ◽  
Roc Curves ◽  
Site Investigation ◽  
Susceptibility Mapping ◽  
Landslide Susceptibility Mapping ◽  
Operating Characteristics ◽  
Statistical Index ◽  
Hierarchy Process

A number of approaches have been used inlandslide susceptibility mappingand there seems to be no agreement on which approach provides best prediction of landslide susceptibility. This study was conducted to develop landslide susceptibility mapsof the Tinalah watershed at a 1:25000 scale using analytical hierarchy process (AHP), statistical index (SI), index of entropy (IOE) and logistic regression (LR) approaches and to compare performance of those approaches in predicting landslide susceptibility. Out of 114 landslides identified during site investigation, 86 landslides were selected for landslide susceptibility analyses, while the remaining 28 landslides were used to verify the results. Factors controlling landslides considered in the landslide susceptibility mapping were slope inclination, lithology, distance to fault, land use, distance to river, and rainfall. Analyses of Receiver Operating Characteristics (ROC) curves showed that the AUC values of the landslide susceptibility maps derived using AHP, SI, IOE and LR approaches were 0.784, 0.688, 0.827 and 0.834, respectively. The LR approach was concluded to perform the best in predicting landslide susceptibility in the study area.

scholarly journals Comparison of Artificial Neural Network with Logistic Regression as Classification Models for Variable Selection for Prediction of Breast Cancer Patient Outcomes

2010 ◽  
Vol 2010 ◽  
pp. 1-11 ◽  
Author(s):  
Valérie Bourdès ◽  
Stéphane Bonnevay ◽  
Paolo Lisboa ◽  
Rémy Defrance ◽  
David Pérol ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Logistic Regression ◽  
Disease Free Survival ◽  
Roc Curves ◽  
Disease Recurrence ◽  
Breast Cancer Patients ◽  
Operating Characteristics ◽  
Free Survival ◽  
Breast Cancer Cohort ◽  
Disease Free

The aim of this study was to compare multilayer perceptron neural networks (NNs) with standard logistic regression (LR) to identify key covariates impacting on mortality from cancer causes, disease-free survival (DFS), and disease recurrence using Area Under Receiver-Operating Characteristics (AUROC) in breast cancer patients. From 1996 to 2004, 2,535 patients diagnosed with primary breast cancer entered into the study at a single French centre, where they received standard treatment. For specific mortality as well as DFS analysis, the ROC curves were greater with the NN models compared to LR model with better sensitivity and specificity. Four predictive factors were retained by both approaches for mortality: clinical size stage, Scarff Bloom Richardson grade, number of invaded nodes, and progesterone receptor. The results enhanced the relevance of the use of NN models in predictive analysis in oncology, which appeared to be more accurate in prediction in this French breast cancer cohort.

scholarly journals Comparison of a Bayesian Network with a Logistic Regression Model to Forecast IgA Nephropathy

2013 ◽  
Vol 2013 ◽  
pp. 1-6 ◽  
Author(s):  
Michel Ducher ◽  
Emilie Kalbacher ◽  
François Combarnous ◽  
Jérome Finaz de Vilaine ◽  
Brigitte McGregor ◽  
...  
Keyword(s):  
Logistic Regression ◽  
Clinical Practice ◽  
Bayesian Network ◽  
Iga Nephropathy ◽  
Logistic Regression Model ◽  
Roc Curves ◽  
Biological Data ◽  
Receiver Operating Characteristics ◽  
Operating Characteristics ◽  
Biological Criteria

Models are increasingly used in clinical practice to improve the accuracy of diagnosis. The aim of our work was to compare a Bayesian network to logistic regression to forecast IgA nephropathy (IgAN) from simple clinical and biological criteria. Retrospectively, we pooled the results of all biopsies(n=155)performed by nephrologists in a specialist clinical facility between 2002 and 2009. Two groups were constituted at random. The first subgroup was used to determine the parameters of the models adjusted to data by logistic regression or Bayesian network, and the second was used to compare the performances of the models using receiver operating characteristics (ROC) curves. IgAN was found (on pathology) in 44 patients. Areas under the ROC curves provided by both methods were highly significant but not different from each other. Based on the highest Youden indices, sensitivity reached (100% versus 67%) and specificity (73% versus 95%) using the Bayesian network and logistic regression, respectively. A Bayesian network is at least as efficient as logistic regression to estimate the probability of a patient suffering IgAN, using simple clinical and biological data obtained during consultation.

Conscious access to fear-relevant information is mediated by threshold

2011 ◽  
Vol 42 (2) ◽  
pp. 56-64 ◽  
Author(s):  
Remigiusz Szczepanowski
Keyword(s):  
Present Report ◽  
Intrinsic Property ◽  
Roc Curves ◽  
Relevant Information ◽  
Perceptual Processing ◽  
High Threshold ◽  
Operating Characteristics ◽  
Emotional Perception ◽  
Fearful Faces ◽  
The Subject

Conscious access to fear-relevant information is mediated by thresholdThe present report proposed a model of access consciousness to fear-relevant information according to which there is a threshold for emotional perception beyond that the subject makes hits with no false alarm. The model was examined by having the participants performed a confidence-ratings masking task with fearful faces. Measures of the thresholds for conscious access were taken by looking at the receiver operating characteristics (ROC) curves generated from a three-state low- and high-threshold (3-LHT) model by Krantz. Indeed, the analysis of the masking data revealed that the ROCs had threshold-like-nature (a two-limb shape) rather continuous (a curvilinear shape) challenging in this fashion the classical signal-detection view on perceptual processing. Moreover, the threshold ROC curve exhibited the specific y-intercepts relevant to conscious access performance. The study suggests that the threshold can be an intrinsic property of conscious access, mediating emotional contents between perceptual states and consciousness.

scholarly journals Haemodynamic assessment of right ventricular overload in pulmonary hypertension: old parameters still fit better than new ones

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
A Pereira ◽  
J.G Santos ◽  
M.J Loureiro ◽  
F Ferreira ◽  
A.R Almeida ◽  
...  
Keyword(s):  
Pulmonary Hypertension ◽  
Logistic Regression ◽  
Adverse Events ◽  
Predictive Value ◽  
Roc Curves ◽  
Multivariate Logistic Regression ◽  
Discrimination Power ◽  
Haemodynamic Parameters ◽  
All Cause Mortality ◽  
Pulmonary Arterial

Abstract Introduction Right ventricular (RV) adaptation to the increased pulmonary load is a key determinant of outcomes in pulmonary hypertension (PH). Pulmonary vascular resistance (PVR) is widely recognized as haemodynamic measure of RV overload. Cardiac filling pressure (CFP), RV stroke work (RVSW), pulmonary artery (PA) compliance and PA pulsatility index (PAPi) are emerging as new haemodynamic parameters to assess RV function. Aims To assess the predictive value of CFP, RVSW, PA compliance and PAPi in PH and to compare it with standard haemodynamic parameters. Methods Retrospective study including all consecutive right heart catheterizations performed from April/2009 to October/2019 in a PH referral centre. Procedures presenting PH were selected [mean pulmonary arterial pressure (mPAP) >20 mmHg, according to the new definition of the 6st World Symposium on PH]. CFP was calculated as [right atrial pressure (RAP) − pulmonary capillary wedge pressure], value >0.63 associated with RV failure; RVSW as CO / [(heart rate × (mPAP-RAP) × 0.0136], value <15; PA compliance [SV / pulmonary arterial systolic pressure (PASP) − pulmonary arterial diastolic pressure (PADP)], value <2.5]; PAPi [(PSAP − PDAP) / RAP, value <1.85]. Multivariate logistic regression was used to identify predictors of all-cause mortality. Receiver operating characteristic (ROC) curves and area under curve (AUC) were used to assess discrimination power. Results From a total of 569 procedures, 470 fulfilled PH criteria: mean age 57.9±16.0 years, 67.7% female, 35.5% performed under pulmonary vasodilator therapy. Pre-capillary PH was diagnosed in 71.9% of cases. Chronic thromboembolic PH was the most common subtype (34.4%). Concerning standard haemodynamic parameters: mPAP was 39.0±12.0 mmHg, mean RAP 8.0±5.0 mmHg, mean RVP 7.5±5.0 uWood and CI 2.5±0.8 L/min/m2. Median value of CFP was 0.6 (IQR 0.4–0.8), RVSW 15.2 (IQR 9.7–25.0), PA compliance 2.1 (IQR 0.9–2.9) and PAPi 5.3 (IQR 3.2–8.5). All-cause mortality rate was 22.8%. Patients experiencing adverse events had lower values of cardiac index (2.3±0.6 vs 2.6±0.8 L/min/m2, p<0.01), RVSW (11.2 vs 16.7, p<0.01) and PA compliance (2.2 vs 2.9, p<0.01) and higher values of PVR (10.0±5.5 versus 6.8±4.6 uWood, p<0.01) and mean RAP (9.9±6.1 versus 7.4±4.5, p<0.01). Multivariate logistic regression identified 2 independent predictors of adverse events: mean RAP (OR 1.08, 95% CI 1.02–1.13, p<0.01) and PVR (OR 1.11, 95% CI 1.06–1.17, p<0.01). According to the ROC curves, new haemodynamic parameters did not have acceptable discrimination power to adverse events occurrence (figure). Conclusions In this study, new haemodynamic parameters to assess RV overload in PH were not independent predictors of adverse events as opposite to standard haemodynamic parameters. Further studies are needed to clarify their predictive value, as it has major implications for understanding the arterial load in diseases of the pulmonary circulation. Funding Acknowledgement Type of funding source: None

scholarly journals Nomogram based on homogeneous and heterogeneous associated factors for predicting distant metastases in patients with colorectal cancer

2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Tianwen Luo ◽  
Yutong Wang ◽  
Xuefeng Shan ◽  
Ye Bai ◽  
Chun Huang ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Risk Factors ◽  
Lung Metastases ◽  
Roc Curves ◽  
Distant Metastases ◽  
Good Prediction ◽  
Operating Characteristics ◽  
Good Prediction Performance ◽  
Different Types ◽  
Associated Risk Factors

Abstract Background The identification of the homogeneous and heterogeneous risk factors for different types of metastases in colorectal cancer (CRC) may shed light on the aetiology and help individualize prophylactic treatment. The present study characterized the incidence differences and identified the homogeneous and heterogeneous risk factors associated with distant metastases in CRC. Methods CRC patients registered in the SEER database between 2010 and 2016 were included in this study. Logistic regression was used to analyse homogeneous and heterogeneous risk factors for the occurrence of different types of metastases. Nomograms were constructed to predict the risk for developing metastases, and the performance was quantitatively assessed using the receiver operating characteristics (ROC) curve and calibration curve. Results A total of 204,595 eligible CRC patients were included in our study, and 17.07% of them had distant metastases. The overall incidences of liver metastases, lung metastases, bone metastases, and brain metastases were 15.34%, 5.22%, 1.26%, and 0.29%, respectively. The incidence of distant metastases differed by age, gender, and the original CRC sites. Poorly differentiated grade, more lymphatic metastasis, higher carcinoembryonic antigen (CEA), and different metastatic organs were all positively associated with four patterns of metastases. In contrast, age, sex, race, insurance status, position, and T stage were heterogeneously associated with metastases. The calibration and ROC curves exhibited good performance for predicting distant metastases. Conclusions The incidence of distant metastases in CRC exhibited distinct differences, and the patients had homogeneous and heterogeneous associated risk factors. Although limited risk factors were included in the present study, the established nomogram showed good prediction performance.

scholarly journals Right choice of a method for determination of cut-off values: A statistical tool for a diagnostic test

2014 ◽  
Vol 5 (3) ◽  
pp. 30-34 ◽  
Author(s):  
Balkishan Sharma ◽  
Ravikant Jain
Keyword(s):  
Logistic Regression ◽  
Discriminant Analysis ◽  
Diagnostic Test ◽  
Roc Curve ◽  
Traditional Method ◽  
Statistical Technique ◽  
Curve Analysis ◽  
Operating Characteristics ◽  
Roc Curve Analysis

Objective: The clinical diagnostic tests are generally used to identify the presence of a disease. The cutoff value of a diagnostic test should be chosen to maximize the advantage that accrues from testing a population of human and others. When a diagnostic test is to be used in a clinical condition, there may be an opportunity to improve the test by changing the cutoff value. To enhance the accuracy of diagnosis is to develop new tests by using a proper statistical technique with optimum sensitivity and specificity. Method: Mean±2SD method, Logistic Regression Analysis, Receivers Operating Characteristics (ROC) curve analysis and Discriminant Analysis (DA) have been discussed with their respective applications. Results: The study highlighted some important methods to determine the cutoff points for a diagnostic test. The traditional method is to identify the cut-off values is Mean±2SD method. Logistic Regression Analysis, Receivers Operating Characteristics (ROC) curve analysis and Discriminant Analysis (DA) have been proved to be beneficial statistical tools for determination of cut-off points.Conclusion: There may be an opportunity to improve the test by changing the cut-off value with the help of a correctly identified statistical technique in a clinical condition when a diagnostic test is to be used. The traditional method is to identify the cut-off values is Mean ± 2SD method. It was evidenced in certain conditions that logistic regression is found to be a good predictor and the validity of the same can be confirmed by identifying the area under the ROC curve. Abbreviations: ROC-Receiver operating characteristics and DA-Discriminant Analysis. Asian Journal of Medical Science, Volume-5(3) 2014: 30-34 http://dx.doi.org/10.3126/ajms.v5i3.9296      

Alternative models and randomization techniques for Bayesian response-adaptive randomization with binary outcomes

2021 ◽  
pp. 174077452110101
Author(s):  
Jennifer Proper ◽  
John Connett ◽  
Thomas Murray
Keyword(s):  
Logistic Regression ◽  
Sample Size ◽  
Error Rate ◽  
Adaptive Design ◽  
Type I Error ◽  
Probability Model ◽  
Binary Outcomes ◽  
Type I ◽  
Operating Characteristics ◽  
Type I Error Rate

Background: Bayesian response-adaptive designs, which data adaptively alter the allocation ratio in favor of the better performing treatment, are often criticized for engendering a non-trivial probability of a subject imbalance in favor of the inferior treatment, inflating type I error rate, and increasing sample size requirements. The implementation of these designs using the Thompson sampling methods has generally assumed a simple beta-binomial probability model in the literature; however, the effect of these choices on the resulting design operating characteristics relative to other reasonable alternatives has not been fully examined. Motivated by the Advanced R2 Eperfusion STrategies for Refractory Cardiac Arrest trial, we posit that a logistic probability model coupled with an urn or permuted block randomization method will alleviate some of the practical limitations engendered by the conventional implementation of a two-arm Bayesian response-adaptive design with binary outcomes. In this article, we discuss up to what extent this solution works and when it does not. Methods: A computer simulation study was performed to evaluate the relative merits of a Bayesian response-adaptive design for the Advanced R2 Eperfusion STrategies for Refractory Cardiac Arrest trial using the Thompson sampling methods based on a logistic regression probability model coupled with either an urn or permuted block randomization method that limits deviations from the evolving target allocation ratio. The different implementations of the response-adaptive design were evaluated for type I error rate control across various null response rates and power, among other performance metrics. Results: The logistic regression probability model engenders smaller average sample sizes with similar power, better control over type I error rate, and more favorable treatment arm sample size distributions than the conventional beta-binomial probability model, and designs using the alternative randomization methods have a negligible chance of a sample size imbalance in the wrong direction. Conclusion: Pairing the logistic regression probability model with either of the alternative randomization methods results in a much improved response-adaptive design in regard to important operating characteristics, including type I error rate control and the risk of a sample size imbalance in favor of the inferior treatment.

Sign in / Sign up

Export Citation Format

Share Document