Conventional Logistic Regression Models Research Articles

Machine-learning-based prediction of cardiovascular events for hyperlipidemia population with lipid variability and remnant cholesterol as biomarkers.

Dyslipidemia poses a significant risk for the progression to cardiovascular diseases. Despite the identification of numerous risk factors and the proposal of various risk scales, there is still an urgent need for effective predictive models for the onset of cardiovascular diseases in the hyperlipidemic population, which are essential for the prevention of CVD. We carried out a retrospective cohort study with 23,548 hyperlipidemia patients in Shenzhen Health Information Big Data Platform, including 11,723 CVD onset cases in a 3-year follow-up. The population was randomly divided into 70% as an independent training dataset and remaining 30% as test set. Four distinct machine-learning algorithms were implemented on the training dataset with the aim of developing highly accurate predictive models, and their performance was subsequently benchmarked against conventional risk assessment scales. An ablation study was also carried out to analyze the impact of individual risk factors to model performance. The non-linear algorithm, LightGBM, excelled in forecasting the incidence of cardiovascular disease within 3years, achieving an area under the 'receiver operating characteristic curve' (AUROC) of 0.883. This performance surpassed that of the conventional logistic regression model, which had an AUROC of 0.725, on identical datasets. Concurrently, in direct comparative analyses, machine-learning approaches have notably outperformed the three traditional risk assessment methods within their respective applicable populations. These include the Framingham cardiovascular disease risk score, 2019 ESC/EAS guidelines for the management of dyslipidemia and the 2016 Chinese recommendations for the management of dyslipidemia in adults. Further analysis of risk factors showed that the variability of blood lipid levels and remnant cholesterol played an important role in indicating an increased risk of CVD. We have shown that the application of machine-learning techniques significantly enhances the precision of cardiovascular risk forecasting among hyperlipidemic patients, addressing the critical issue of disease prediction's heterogeneity and non-linearity. Furthermore, some recently-suggested biomarkers, including blood lipid variability and remnant cholesterol are also important predictors of cardiovascular events, suggesting the importance of continuous lipid monitoring and healthcare profiling through big data platforms.

Comparing AI/ML approaches and classical regression for predictive modeling using large population health databases: Applications to COVID-19 case prediction

BackgroundResearch comparing artificial intelligence and machine learning (AI/ML) methods with classical statistical methods applied to large population health databases is limited. ObjectivesThis retrospective cohort study aimed to compare the predictive performance of AI/ML algorithms against conventional multivariate logistic regression models using linked health administrative data. MethodsUsing Ontario's population health databases, we created a cohort of residents of the city of Ottawa, Ontario, who underwent a PCR test for COVID-19 between March 10, 2020, and May 13, 2021. Using demographic, socio-economic and health data (including COVID-19 PCR test results and available, symptom data), we developed predictive models for the purpose of COVID-19 case identification using the following approaches: classical multivariate logistic regression (LR); deep neural network (DNN); random forest (RF); and gradient boosting trees (GBT). Model performance comparisons were made using the area under the curve (AUC) swarm plot for 10-fold cross-validation. ResultsThe cohort consisted of n = 351,248 Ottawa residents tested for COVID-19 during the study period. Among whom, a total of n = 883,879 unique COVID-19 tests were performed (2.6 % positive test results). Inclusion of COVID-19 symptoms data in the analysis improved model performance and variable predictive value across all tested models (p < 0.0001), with the 10-fold cross-validation AUC increasing to near or over 0.7 in all models when symptoms data were included. In various pairwise comparisons, the GBT method had the highest predictive ability (AUC = 0.796 ± 0.017), significantly outperforming multivariate logistic regression and the other AI/ML approaches. ConclusionsConventional multivariate regression-based models are better than some and worse than other machine learning algorithms to provide good predictive accuracy in a moderate dataset with a reasonable number of features. However, whenever possible, the AI/ML GBT approach should be considered.

Geographically weighted logistic regression model for identifying risk factors for malaria infection among under-5 children in Ghana

Malaria remains a significant public health challenge globally, particularly affecting children under-5 years of age due to their underdeveloped immune systems. Identifying the risk factors associated with malaria infection in this vulnerable group is crucial for improving policy formulation and creating effective training programs. However, there is limited information on how the relationship between malaria risk and associated factors varies across different regions, especially among children in Ghana. This is important because understanding these spatial variations can enhance targeted interventions including area remediation and resource allocation. To address this gap, a geographically weighted logistic regression (GWLR) model was developed to identify spatially varying risk factors for malaria infection among children under five in Ghana. The model was built on the premise that the relationship between malaria and potential risk factors is not uniform across geographic areas. Data from the Ghana Malaria Indicator Survey collected through the demographic and health survey program were used for analysis. The study found that the GWLR model fit the data better than the conventional binary logistic regression (BLR) model, based on the information criterion used and mode evaluation metrics. The results indicated that risk factors for malaria such as a child's age, anaemia status, dwellings sprayed, place of residence, electricity access, NHIS (National Health Insurance Scheme) coverage, age of the household head, and household wealth index, were non-stationary across the study area. These findings underscore the importance of spatially tailored interventions to reduce malaria risk in children under-5. The results contribute to the body of literature on malaria risk factors and provide valuable insights for Ghana's national malaria control policies, potentially enhancing the effectiveness of future public health strategies.

Machine learning model predicts airway stenosis requiring clinical intervention in patients after lung transplantation: a retrospective case-controlled study

BackgroundPatients with airway stenosis (AS) are associated with considerable morbidity and mortality after lung transplantation (LTx). This study aims to develop and validate machine learning (ML) models to predict AS requiring clinical intervention in patients after LTx.MethodsPatients who underwent LTx between January 2017 and December 2019 were reviewed. The conventional logistic regression (LR) model was fitted by the independent risk factors which were determined by multivariate LR. The optimal ML model was determined based on 7 feature selection methods and 8 ML algorithms. Model performance was assessed by the area under the curve (AUC) and brier score, which were internally validated by the bootstrap method.ResultsA total of 381 LTx patients were included, and 40 (10.5%) patients developed AS. Multivariate analysis indicated that male, pulmonary arterial hypertension, and postoperative 6-min walking test were significantly associated with AS (all P < 0.001). The conventional LR model showed performance with an AUC of 0.689 and brier score of 0.091. In total, 56 ML models were developed and the optimal ML model was the model fitted using a random forest algorithm with a determination coefficient feature selection method. The optimal model exhibited the highest AUC and brier score values of 0.760 (95% confidence interval [CI], 0.666–0.864) and 0.085 (95% CI, 0.058–0.117) among all ML models, which was superior to the conventional LR model.ConclusionsThe optimal ML model, which was developed by clinical characteristics, allows for the satisfactory prediction of AS in patients after LTx.

Prediction of 28-Day All-Cause Mortality in Heart Failure Patients with Clostridioides difficile Infection Using Machine Learning Models: Evidence from the MIMIC-IV Database.

Heart failure (HF) may induce bowel hypoperfusion, leading to hypoxia of the villa of the bowel wall and the occurrence of Clostridioides difficile infection (CDI). However, the risk factors for the development of CDI in HF patients have yet to be fully illustrated, especially because of a lack of evidence from real-world data. Clinical data and survival situations of HF patients with CDI admitted to ICU were extracted from the Medical Information Mart for Intensive Care (MIMIC)-IV database. For developing a model that can predict 28-day all-cause mortality in HF patients with CDI, the Recursive Feature Elimination with Cross-Validation (RFE-CV) method was used for feature selection. And nine machine learning (ML) algorithms, including logistic regression (LR), decision tree, Bayesian, adaptive boosting, random forest (RF), gradient boosting decision tree, XGBoost, light gradient boosting machine, and categorical boosting, were applied for model construction. After training and hyperparameter optimization of the models through grid search 5-fold cross-validation, the performance of models was evaluated by the area under curve (AUC), accuracy, sensitivity, specificity, precision, negative predictive value, and F1 score. Furthermore, the SHapley Additive exPlanations (SHAP) method was used to interpret the optimal model. A total of 526 HF patients with CDI were included in the study, of whom 99 cases (18.8%) experienced death within 28 days. Eighteen of the 57 variables were selected for the model construction algorithm for model construction. Among the ML models considered, the RF model emerged as the optimal model achieving the accuracy, F1-score, and AUC values of 0.821, 0.596, and 0.864, respectively. The net benefit of the model surpassed other models at 16%-22% threshold probabilities based on decision curve analysis. According to the importance of features in the RF model, red blood cell distribution width, blood urea nitrogen, Simplified Acute Physiology Score II, Sequential Organ Failure Assessment, and white blood cell count were highlighted as the five most influential variables. We developed ML models to predict 28-day all-cause mortality in HF patients associated with CDI in the ICU, which are more effective than the conventional LR model. The RF model has the best performance among all the ML models employed. It may be useful to help clinicians identify high-risk HF patients with CDI.

Minimizing confounding in comparative observational studies with time-to-event outcomes: An extensive comparison of covariate balancing methods using Monte Carlo simulation.

Observational studies are frequently used in clinical research to estimate the effects of treatments or exposures on outcomes. To reduce the effects of confounding when estimating treatment effects, covariate balancing methods are frequently implemented. This study evaluated, using extensive Monte Carlo simulation, several methods of covariate balancing, and two methods for propensity score estimation, for estimating the average treatment effect on the treated using a hazard ratio from a Cox proportional hazards model. With respect to minimizing bias and maximizing accuracy (as measured by the mean square error) of the treatment effect, the average treatment effect on the treated weighting, fine stratification, and optimal full matching with a conventional logistic regression model for the propensity score performed best across all simulated conditions. Other methods performed well in specific circumstances, such as pair matching when sample sizes were large (n = 5000) and the proportion treated was < 0.25. Statistical power was generally higher for weighting methods than matching methods, and Type I error rates were at or below the nominal level for balancing methods with unbiased treatment effect estimates. There was also a decreasing effective sample size with an increasing number of strata, therefore for stratification-based weighting methods, it may be important to consider fewer strata. Generally, we recommend methods that performed well in our simulations, although the identification of methods that performed well is necessarily limited by the specific features of our simulation. The methods are illustrated using a real-world example comparing beta blockers and angiotensin-converting enzyme inhibitors among hypertensive patients at risk for incident stroke.

Ultrasound-based radiomics machine learning models for diagnosing cervical lymph node metastasis in patients with non-small cell lung cancer: a multicentre study

BackgroundCervical lymph node metastasis (LNM) is an important prognostic factor for patients with non-small cell lung cancer (NSCLC). We aimed to develop and validate machine learning models that use ultrasound radiomic and descriptive semantic features to diagnose cervical LNM in patients with NSCLC.MethodsThis study included NSCLC patients who underwent neck ultrasound examination followed by cervical lymph node (LN) biopsy between January 2019 and January 2022 from three institutes. Radiomic features were extracted from the ultrasound images at the maximum cross-sectional areas of cervical LNs. Logistic regression (LR) and random forest (RF) models were developed. Model performance was assessed by the area under the curve (AUC) and accuracy, validated internally and externally by fivefold cross-validation and hold-out method, respectively.ResultsIn total, 313 patients with a median age of 64 years were included, and 276 (88.18%) had cervical LNM. Three descriptive semantic features, including long diameter, shape, and corticomedullary boundary, were selected by multivariate analysis. Out of the 474 identified radiomic features, 9 were determined to fit the LR model, while 15 fit the RF model. The average AUCs of the semantic and radiomics models were 0.876 (range: 0.781–0.961) and 0.883 (range: 0.798–0.966), respectively. However, the average AUC was higher for the semantic-radiomics combined LR model (0.901; range: 0.862–0.927). When the RF algorithm was applied, the average AUCs of the radiomics and semantic-radiomics combined models were improved to 0.908 (range: 0.837–0.966) and 0.922 (range: 0.872–0.982), respectively. The models tested by the hold-out method had similar results, with the semantic-radiomics combined RF model achieving the highest AUC value of 0.901 (95% CI, 0.886–0.968).ConclusionsThe ultrasound radiomic models showed potential for accurately diagnosing cervical LNM in patients with NSCLC when integrated with descriptive semantic features. The RF model outperformed the conventional LR model in diagnosing cervical LNM in NSCLC patients.

Bayesian compositional models for ordinal response.

Ordinal response is commonly found in medicine, biology, and other fields. In many situations, the predictors for this ordinal response are compositional, which means that the sum of predictors for each sample is fixed. Examples of compositional data include the relative abundance of species in microbiome data and the relative frequency of nutrition concentrations. Moreover, the predictors that are strongly correlated tend to have similar influence on the response outcome. Conventional cumulative logistic regression models for ordinal responses ignore the fixed-sum constraint on predictors and their associated interrelationships, and thus are not appropriate for analyzing compositional predictors.To solve this problem, we proposed Bayesian Compositional Models for Ordinal Response to analyze the relationship between compositional data and an ordinal response with a structured regularized horseshoe prior for the compositional coefficients and a soft sum-to-zero restriction on coefficients through the prior distribution. The method was implemented with R package rstan using efficient Hamiltonian Monte Carlo algorithm. We performed simulations to compare the proposed approach and existing methods for ordinal responses. Results revealed that our proposed method outperformed the existing methods in terms of parameter estimation and prediction. We also applied the proposed method to a microbiome study HMP2Data, to find microorganisms linked to ordinal inflammatory bowel disease levels. To make this work reproducible, the code and data used in this paper are available at https://github.com/Li-Zhang28/BCO.

Compare deep learning model and conventional logistic regression model for the identification of unstable saccular intracranial aneurysms in computed tomography angiography.

It is crucial to distinguish unstable from stable intracranial aneurysms (IAs) as early as possible to derive optimal clinical decision-making for further treatment or follow-up. The aim of this study was to investigate the value of a deep learning model (DLM) in identifying unstable IAs from computed tomography angiography (CTA) images and to compare its discriminatory ability with that of a conventional logistic regression model (LRM). From August 2011 to May 2021, a total of 1,049 patients with 681 unstable IAs and 556 stable IAs were retrospectively analyzed. IAs were randomly divided into training (64%), internal validation (16%), and test sets (20%). Convolutional neural network (CNN) analysis and conventional logistic regression (LR) were used to predict which IAs were unstable. The area under the curve (AUC), sensitivity, specificity and accuracy were calculated to evaluate the discriminating ability of the models. One hundred and ninety-seven patients with 229 IAs from Banan Hospital were used for external validation sets. The conventional LRM showed 11 unstable risk factors, including clinical and IA characteristics. The LRM had an AUC of 0.963 [95% confidence interval (CI): 0.941-0.986], a sensitivity, specificity and accuracy on the external validation set of 0.922, 0.906, and 0.913, respectively, in predicting unstable IAs. In predicting unstable IAs, the DLM had an AUC of 0.771 (95% CI: 0.582-0.960), a sensitivity, specificity and accuracy on the external validation set of 0.694, 0.929, and 0.782, respectively. The CNN-based DLM applied to CTA images did not outperform the conventional LRM in predicting unstable IAs. The patient clinical and IA morphological parameters remain critical factors for ensuring IA stability. Further studies are needed to enhance the diagnostic accuracy.

LightGBM outperforms other machine learning techniques in predicting graft failure after liver transplantation: Creation of a predictive model through large-scale analysis.

The incidence of graft failure following liver transplantation (LTx) is consistent. While traditional risk scores for LTx have limited accuracy, the potential of machine learning (ML) in this area remains uncertain, despite its promise in other transplant domains. This study aims to determine ML's predictive limitations in LTx by replicating methods used in previous heart transplant research. This study utilized the UNOS STAR database, selecting 64,384 adult patients who underwent LTx between 2010 and 2020. Gradient boosting models (XGBoost and LightGBM) were used to predict 14, 30, and 90-day graft failure compared to conventional logistic regression model. Models were evaluated using both shuffled and rolling cross-validation (CV) methodologies. Model performance was assessed using the AUC across validation iterations. In a study comparing predictive models for 14-day, 30-day and 90-day graft survival, LightGBM consistently outperformed other models, achieving the highest AUC of.740,.722, and.700 in shuffled CV methods. However, in rolling CV the accuracy of the model declined across every ML algorithm. The analysis revealed influential factors for graft survival prediction across all models, including total bilirubin, medical condition, recipient age, and donor AST, among others. Several features like donor age and recipient diabetes history were important in two out of three models. LightGBM enhances short-term graft survival predictions post-LTx. However, due to changing medical practices and selection criteria, continuous model evaluation is essential. Future studies should focus on temporal variations, clinical implications, and ensure model transparency for broader medical utility.

Machine learning models for predicting in-hospital outcomes after non-surgical treatment among patients with moderate-to-severe traumatic brain injury

AimThis study aims to develop prediction models for in-hospital outcomes after non-surgical treatment among patients with moderate-to-severe traumatic brain injury (TBI). MethodWe conducted a retrospective review of patients hospitalized for moderate-to-severe TBI in our department from 2011 to 2020. Five machine learning (ML) algorithms and the conventional logistic regression (LR) model were employed to predict in-hospital mortality and the Glasgow Outcome Scale (GOS) functional outcomes. These models utilized clinical and routine blood data collected upon admission. ResultsThis study included a total of 196 patients who received only non-surgical treatment after moderate-to-severe TBI. When predicting mortality, ML models achieved area under the curve (AUC) values of 0.921 to 0.994 using clinical and routine blood data, and 0.877 to 0.982 using only clinical data. In comparison, LR models yielded AUCs of 0.762 and 0.730 respectively. When predicting the GOS outcome, ML models achieved AUCs of 0.870 to 0.915 using clinical and routine blood data, and 0.858 to 0.927 using only clinical data. In comparison, the LR model yielded AUCs of 0.798 and 0.787 respectively. Repeated internal validation showed that the contributions of routine blood data for prediction models may depend on different prediction algorithms and different outcome measurements. ConclusionThe study reported ML-based prediction models that provided rapid and accurate predictions on short-term outcomes after non-surgical treatment among patients with moderate-to-severe TBI. The study also highlighted the superiority of ML models over conventional LR models and proposed the complex contributions of routine blood data in such predictions.

Prediction of Major Adverse Cardiac Events After Transcatheter Aortic Valve Implantation: A Machine Learning Approach with GRACE Score.

Predictive risk scores have a significant impact on patient selection and assessing the likelihood of complications following interventions in patients with severe aortic stenosis (AS). This study aims to explore the utility of machine learning (ML) techniques in predicting 30-day major adverse cardiac events (MACE) by analyzing parameters, including the Global Registry of Acute Coronary Events (GRACE) score. This retrospective, multi-center, observational study enrolled 453 consecutive patients diagnosed with severe AS who underwent transcatheter aortic valve implantation (TAVI) from April 2020 to January 2023. The primary outcome was defined as a composition of MACE comprising periprocedural myocardial infarction (MI), cerebrovascular events (CVE), and all-cause mortality during the 1-month follow-up period after the procedure. Conventional binomial logistic regression and ML models were utilized and compared for prediction purposes. The study population had a mean age of 76.1, with 40.8% being male. The primary endpoint was observed in 7.5% of cases. Among the individual components of the primary endpoint, the rates of all-cause mortality, MI, and CVE were reported as 4.2%, 2.4%, and 1.9%, respectively. The ML-based Extreme Gradient Boosting (XGBoost) model with the GRACE score demonstrated superior discriminative performance in predicting the primary endpoint, compared to both the ML model without the GRACE score and the conventional regression model [Area Under the Curve (AUC)= 0.98 (0.91-0.99), AUC= 0,87 (0.80-0.98), AUC= 0.84 (0.79-0.96)]. ML techniques hold the potential to enhance outcomes in clinical practice, especially when utilized alongside established clinical tools such as the GRACE score.

RADT-41. PREDICTION OF TREATMENT OUTCOME IN GAMMA KNIFE RADIOSURGERY OF BENIGN MENINGIOMAS

Abstract Objective This study aimed to assess the predictability of machine learning (ML) ensemble models for the success of Gamma Knife radiosurgery (GKRS) on non-skull base benign meningiomas. METHODS A total of 331 non-skull base benign meningiomas treated primarily with GKRS between 1997 and 2015 were retrospectively analyzed. When a tumor remained not larger than 125% of the initial volume at the last imaging follow-up and no significant change in the high signal intensity area around the tumor, it was classified as 'treatment success (TS).' Univariate correlation and multivariate logistic regression analyses were performed to identify statistically significant prognostic factors. Four ensemble algorithms were trained to predict the outcome of GKRS. The importance of features related to TS was analyzed. RESULTS The median MR imaging follow-up period was 8.6 years. The tumor control and signal change rates were 91.5% and 16.3%, respectively, and the TS rate was 77.6%. The conventional multivariate logistic regression model showed 83.7% accuracy and 37.8% specificity. The ML models had accuracies between 78.0% and 82.2% and specificities between 30.3% and 80.7%, and they were similar to those of conventional logistic regression analyses. The smaller brain volume absorbed more than 14 Gy, and the absence of signal change before GKRS was a favorable prognostic factor for TS in all models. The bagging algorithm with selected features showed the best specificity of 80.7% and the most significant AUC of 0.723. CONCLUSION A bagging ensemble algorithm with selected features had better specificity with similar accuracy than conventional multivariate logistic regression in predicting GKRS outcomes for non-skull base meningiomas.

Identifying the severity of diabetic retinopathy by visual function measures using both traditional statistical methods and interpretable machine learning: a cross-sectional study

Aims/hypothesisTo determine the extent to which diabetic retinopathy severity stage may be classified using machine learning (ML) and commonly used clinical measures of visual function together with age and sex.MethodsWe measured the visual function of 1901 eyes from 1032 participants in the Northern Ireland Sensory Ageing Study, deriving 12 variables from nine visual function tests. Missing values were imputed using chained equations. Participants were divided into four groups using clinical measures and grading of ophthalmic images: no diabetes mellitus (no DM), diabetes but no diabetic retinopathy (DM no DR), diabetic retinopathy without diabetic macular oedema (DR no DMO) and diabetic retinopathy with DMO (DR with DMO). Ensemble ML models were fitted to classify group membership for three tasks, distinguishing (A) the DM no DR group from the no DM group; (B) the DR no DMO group from the DM no DR group; and (C) the DR with DMO group from the DR no DMO group. More conventional multiple logistic regression models were also fitted for comparison. An interpretable ML technique was used to rank the contribution of visual function variables to predictions and to disentangle associations between diabetic eye disease and visual function from artefacts of the data collection process.ResultsThe performance of the ensemble ML models was good across all three classification tasks, with accuracies of 0.92, 1.00 and 0.84, respectively, for tasks A–C, substantially exceeding the accuracies for logistic regression (0.84, 0.61 and 0.80, respectively). Reading index was highly ranked for tasks A and B, whereas near visual acuity and Moorfields chart acuity were important for task C. Microperimetry variables ranked highly for all three tasks, but this was partly due to a data artefact (a large proportion of missing values).Conclusions/interpretationEnsemble ML models predicted status of diabetic eye disease with high accuracy using just age, sex and measures of visual function. Interpretable ML methods enabled us to identify profiles of visual function associated with different stages of diabetic eye disease, and to disentangle associations from artefacts of the data collection process. Together, these two techniques have great potential for developing prediction models using untidy real-world clinical data.Graphical

Effect of alcohol consumption on breast cancer: probabilistic bias analysis for adjustment of exposure misclassification bias and confounders

PurposeThis study was conducted to evaluate the effect of alcohol consumption on breast cancer, adjusting for alcohol consumption misclassification bias and confounders.MethodsThis was a case-control study of 932 women with breast cancer and 1000 healthy control. Using probabilistic bias analysis method, the association between alcohol consumption and breast cancer was adjusted for the misclassification bias of alcohol consumption as well as a minimally sufficient set of adjustment of confounders derived from a causal directed acyclic graph. Population attributable fraction was estimated using the Miettinen’s Formula.ResultsBased on the conventional logistic regression model, the odds ratio estimate between alcohol consumption and breast cancer was 1.05 (95% CI: 0.57, 1.91). However, the adjusted estimates of odds ratio based on the probabilistic bias analysis ranged from 1.82 to 2.29 for non-differential and from 1.93 to 5.67 for differential misclassification. Population attributable fraction ranged from 1.51 to 2.57% using non-differential bias analysis and 1.54–3.56% based on differential bias analysis.ConclusionA marked measurement error was in self-reported alcohol consumption so after correcting misclassification bias, no evidence against independence between alcohol consumption and breast cancer changed to a substantial positive association.

Mathematical Approach in Image Classification using Regression

This study presents a comprehensive evaluation of logistic regression in contrast to a hybrid model combining VGG16 with logistic regression for image classification tasks. The research findings illuminate a striking performance disparity between these two approaches, shedding light on the profound impact of integrating deep learning techniques into image classification. The transition from logistic regression to the VGG16-based hybrid model marks a notable turning point in our investigation. The VGG16 architecture, renowned for its prowess as a feature extractor, showcases an impressive 53.33% surge in accuracy compared to the conventional logistic regression model. This substantial leap underscores the model's capacity to decipher complex image characteristics that elude traditional statistical methods. Furthermore, precision, a crucial metric in classification tasks, experiences a substantial 53% augmentation when adopting the VGG16 hybrid approach. This enhancement signifies the hybrid model's ability to minimize false positives, making it particularly valuable in scenarios where precision holds paramount importance. Equally noteworthy is the substantial 54% improvement observed in both recall and F1-score, emphasizing the VGG16 hybrid model's remarkable capacity to identify and retrieve a higher proportion of true positives while maintaining a balance between precision and recall. This not only amplifies the model's ability to correctly classify images but also mitigates the risk of overlooking relevant instances. These compelling findings underscore the critical role of deep learning, specifically convolutional neural networks (CNNs), in the realm of image classification. The utilization of CNNs, exemplified by the VGG16 architecture, emerges as a game-changer, enabling the capture of intricate image features and patterns that traditional logistic regression struggles to discern. Generally, this study advocates for the integration of advanced deep learning techniques, like VGG16, in image classification endeavors. The substantial performance gains witnessed in accuracy, precision, recall, and F1-score reinforce the pivotal role of convolutional neural networks in enhancing the effectiveness of image classification tasks. By harnessing the power of deep learning, we unlock new horizons in image analysis, paving the way for more accurate and efficient classification systems

Modeling pedestrian behavior at urban signalised intersections using statistical-ANN hybrid approach – Case study of New Delhi

Traffic crashes at intersections are major hazard all over the world. The victims of road crashes are mostly pedestrians. A large share of accidents can be related to unsafe crossing behavior by pedestrians. The accidents are mostly fatal for pedestrians especially if the unsafe act is signal violation. This study primarily focused on cognizing factors which are significantly affecting the signal violation behavior of pedestrians’ at urban signalised intersections. Data is collected at 11 signalised intersections in New Delhi, India using video recording technique. Significant variables are identified using conventional statistics. The variables are then used as input neuron for the Artificial Neural Network (ANN) for modeling signal violation behavior. Sensitivity analysis is performed to identify the relative importance of factors related to the pedestrian signal violation. Host of different factors such as pedestrian demographic, behavioral attributes, crossing state and mobile usage are notably influencing signal violations. Pedestrian crossing speed, crossing path and waiting time are the top three predictors of violation behavior. The accuracy of ANN to predict signal violation behaviour is found to be about 85% and is considerably higher than conventional binary logistic regression (BLR) model. The area under the receiver operating characteristic (ROC) curve (AUC) is found to be 0.753, suggesting good model performance. The results suggest that ANN is a robust alternative to conventional regression, for studying and modeling pedestrian behavior. The findings from this study would assist engineers and policy makers to take proactive measures for designing pedestrian friendly facilities to reduce signal violations. This would ultimately improve pedestrian safety at urban signalised intersections.

Predicting hyperkalemia in patients with advanced chronic kidney disease using the XGBoost model

BackgroundHyperkalemia is a common complication of chronic kidney disease (CKD). Hyperkalemia is associated with mortality, CKD progression, hospitalization, and high healthcare costs in patients with CKD. We developed a machine learning model to predict hyperkalemia in patients with advanced CKD at an outpatient clinic.MethodsThis retrospective study included 1,965 advanced CKD patients between January 1, 2010, and December 31, 2020 in Taiwan. We randomly divided all patients into the training (75%) and testing (25%) datasets. The primary outcome was to predict hyperkalemia (K+ > 5.5 mEq/L) in the next clinic vist. Two nephrologists were enrolled in a human-machine competition. The area under the receiver operating characteristic curves (AUCs), sensitivity, specificity, and accuracy were used to evaluate the performance of XGBoost and conventional logistic regression models with that of these physicians.ResultsIn a human-machine competition of hyperkalemia prediction, the AUC, PPV, and accuracy of the XGBoost model were 0.867 (95% confidence interval: 0.840–0.894), 0.700, and 0.933, which was significantly better than that of our clinicians. There were four variables that were chosen as high-ranking variables in XGBoost and logistic regression models, including hemoglobin, the serum potassium level in the previous visit, angiotensin receptor blocker use, and calcium polystyrene sulfonate use.ConclusionsThe XGBoost model provided better predictive performance for hyperkalemia than physicians at the outpatient clinic.

Using Machine Learning to Predict Abnormal Carotid Intima-Media Thickness in Type 2 Diabetes.

Carotid intima-media thickness (c-IMT) is a reliable risk factor for cardiovascular disease risk in type 2 diabetes (T2D) patients. The present study aimed to compare the effectiveness of different machine learning methods and traditional multiple logistic regression in predicting c-IMT using baseline features and to establish the most significant risk factors in a T2D cohort. We followed up with 924 patients with T2D for four years, with 75% of the participants used for model development. Machine learning methods, including classification and regression tree, random forest, eXtreme gradient boosting, and Naïve Bayes classifier, were used to predict c-IMT. The results showed that all machine learning methods, except for classification and regression tree, were not inferior to multiple logistic regression in predicting c-IMT in terms of higher area under receiver operation curve. The most significant risk factors for c-IMT were age, sex, creatinine, body mass index, diastolic blood pressure, and duration of diabetes, sequentially. Conclusively, machine learning methods could improve the prediction of c-IMT in T2D patients compared to conventional logistic regression models. This could have crucial implications for the early identification and management of cardiovascular disease in T2D patients.

Identifying the Local Impacts of Speed-Related Factors on Tunnel Entrance Crash Severity

According to crash statistics, freeway tunnel entrance zones have higher injury severity rates than other freeway sections. The concentration of traffic injuries is a combination of high speed and low driver alertness. The government of South Korea has deployed the variable speed limit (VSL) as a safety countermeasure to mitigate the consequences of tunnel entrance crashes. This measure has not been quantitatively reviewed, therefore, the effectiveness of implementing VSL at tunnel entrance zones is unknown. The aim of this was to predict the injury severity of crashes occurring at the tunnel entrance zone given the nature of the crash, the size of the tunnel, the roadway geometric characteristics, and environmental factors. It was anticipated that a statistical model calibrated with historical crash data and tunnel features would be able to justify the current locations of VSL implementation and prioritize new VSL implementations. A geographically weighted logistic regression was employed and benchmarked by the conventional logistic regression model, considering the varying geographic locations of freeway tunnel entrances. The geographically weighted logistic regression model presented consistent findings and a slightly better statistical goodness-of-fit than the conventional logistic model. Moreover, the geographically weighted regression captured the local impact of low-level traffic speed on crash injury severities. The findings in this study could help researchers and practitioners evaluate site-specific traffic safety improvement strategies.