Logistic Decision Research Articles

Clinical validation and optimization of machine learning models for early prediction of sepsis

IntroductionSepsis is a global health threat that has a high incidence and mortality rate. Early prediction of sepsis onset can drive effective interventions and improve patients’ outcome.MethodsData were collected retrospectively from a cohort of 2,329 adult patients with positive bacteria cultures from a tertiary hospital in China between October 1, 2019 and September 30, 2020. Thirty six clinical features were selected as inputs for the models. We trained models in predicting sepsis by machine learning (ML) methods, including logistic regression, decision tree, random forest (RF), multi-layer perceptron, and light gradient boosting. We evaluated the performance of the five ML models and the evaluation metrics were: area under the ROC curve (AUC), accuracy, F1-score, sensitivity and specificity. The data of another cohort of 2,286 patients between October 1, 2020 and April 1, 2022 were used to validate the performance of the model performing best in the in the internal validation set. Shapley additive explanations (SHAP) method was applied to evaluate feature importance and explain the predictions of this model.ResultsOf the five machine learning models developed, the RF model demonstrated the best performance in terms of AUC (0.818), F1 value (0.38), and sensitivity (0.746). The RF model also has a comparable AUC (0.771) in the external validation set. The SHAP method identified procalcitonin, albumin, prothrombin time, and sex as the important variables contributing to the prediction of sepsis.DiscussionThe RF model we developed showed the greatest potential for early prediction of sepsis in admitted patients, which could aid clinicians in their decision-making process. Our findings also suggested that male patients with bacterial infections and high procalcitonin levels, lower albumin levels, or prolonged prothrombin times were more likely to develop sepsis.

Analysis of risk factors for co-morbid anxiety and depression in pregnant women.

Co-morbid anxiety and depression (CAD) is defined as the co-existence of anxiety and depression. During pregnancy, women are more prone to negative emotions, such as anxiety and depression, than the general female population. The incidence of CAD during pregnancy is 1-26 %. The aims of this study were to investigate the incidence and influencing factors of CAD during pregnancy. The study cohort included 3053 pregnant women who underwent maternity check-ups at a tertiary hospital in China. Demographic characteristics, level of social support, and psychological characteristics were collected from participants via a self-reported questionnaire, which included the General Demographic Information Questionnaire, Pregnancy Stress Scale, the Generalized Anxiety Scale, Patient Health Questionnaire-9, and Perceived Social Support Scale. A binary logistic regression model and a categorical decision tree based on the Chi-square Automatic Interaction Detector algorithm were used to identify factors influencing CAD during pregnancy, and the differences between the two models were analysed and compared. The results of the logistic regression and decision tree models identified pregnancy stress, social support, pregnancy knowledge, couple relationship satisfaction, advanced maternal age, and occupation as factors influencing CAD during pregnancy. Pregnancy stress was the most influential factor. The areas under the curve of the classification decision tree and logistic regression models were 0.801 (95 % CI: 0.778-0.823) and 0.827 (95 % CI: 0.807-0.847), respectively, with specificities of 63 and 77 %, and sensitivities of 83.9 and 76.3 %. Logistic regression excels when dealing with linear relationships, while decision trees are particularly useful when dealing with nonlinear relationships. Therefore, combining logistic regression and decision trees can achieve some degree of model diversity, thus improving predictive power and confidence in the results.

Using supervised machine learning algorithms to predict bovine leukemia virus seropositivity in dairy cattle in Florida: A 10-year retrospective study

Supervised machine-learning (SML) algorithms are potentially powerful tools that may be used for screening cows for infectious diseases such as bovine leukemia virus (BLV) infection. Here, we compared six different SML algorithms to identify the most important risk factors for predicting BLV seropositivity in dairy cattle in Florida. We used a dataset of 1,279 dairy blood sample records from the Bronson Animal Disease Diagnostic Laboratory that were submitted for BLV antibody testing from 2012 to 2022. The SML algorithms that we used were logistic regression (LR), decision tree (DT), gradient boosting (GB), random forest (RF), neural network (NN), and support vector machine (SVM). A total of 312 serum samples were positive for BLV with corrected seroprevalence of 26.0%. Subject to limitations of the analyzed retrospective data, the RF model was the best model for predicting BLV seropositivity in dairy cattle indicated by the highest Kolmogorov-Smirnov (KS) statistic of 0.75, area under the receiver operating characteristic (AUROC) of 0.93, gain of 2.6; and lowest misclassification rate of 0.10. The LR model was the worst. The RF model showed that the best predictors for BLV seropositivity were age (dairy cows of age ≥ 5 years) and geographic location (southern Florida). We concluded that the RF and other SML algorithms hold promise for predicting BLV seropositivity in dairy cattle and that dairy cattle 5 years of age or older raised in southern Florida have a higher likelihood of testing positive for BLV. This study makes an important methodological contribution to the needed development of predictive tools for effective screening for BLV infection and emphasizes the importance of collecting and using representative data in such predictive models.

Quantitative abnormalities in the β-region of the electrophoretic profile of serum proteins as predictive markers of monoclonality: Machine learning for monoclonality prediction.

To develop a machine learning algorithm aimed at predicting the presence of a monoclonal (M-) protein when the β-globulin fraction is elevated. Patients were selected as part of the University Hospital Laboratory of Brussels routine diagnostic procedures from October 2021 to April 2022. Adult patients with serum protein electrophoresis showing elevated β1 and/or β2 fractions were included. The selection was done following strict exclusion criteria such as acute inflammation, iron deficiency anemia signs or supposed liver disease. To construct a predictive model for prediction of positive immunofixation (IFE) for monoclonality, the following factors were used: age, sex, β1 and β2 concentration (g/L), total proteins (g/L), IgA, IgM, IgG values (g/L) and hypogammaglobulinemia. The dataset underwent a random split, divided into a foundational training set (80%, 247 samples) and a foundational test set (20%, 62 samples). The training sets were subjected to five different algorithms: logistic regression, decision tree, random forest, gradient boosting, and support vector. 309 patients were selected; 149 exhibited a negative IFE and 160 a positive IFE for monoclonality. The evaluation of the five tested models demonstrated very good performance, the chosen model was Random Forest for its high sensitivity (85%) and area under the receiver operating characteristic curve (91%). An accurate algorithm was achieved for predicting the presence of M protein when the β-globulin fraction is elevated which enables early and improved diagnosis of monoclonal gammopathy.

Fraud Detection: A Hybrid Approach With Logistic Regression, Decision Tree, and Random Forest

Fraud Detection: A Hybrid Approach With Logistic Regression, Decision Tree, and Random Forest

RADAR-AD: assessment of multiple remote monitoring technologies for early detection of Alzheimer’s disease

BackgroundAlzheimer’s disease (AD) is a progressive neurodegenerative disorder affecting millions worldwide, leading to cognitive and functional decline. Early detection and intervention are crucial for enhancing the quality of life of patients and their families. Remote Monitoring Technologies (RMTs) offer a promising solution for early detection by tracking changes in behavioral and cognitive functions, such as memory, language, and problem-solving skills. Timely detection of these symptoms can facilitate early intervention, potentially slowing disease progression and enabling appropriate treatment and care.MethodsThe RADAR-AD study was designed to evaluate the accuracy and validity of multiple RMTs in detecting functional decline across various stages of AD in a real-world setting, compared to standard clinical rating scales. Our approach involved a univariate analysis using Analysis of Covariance (ANCOVA) to analyze individual features of six RMTs while adjusting for variables such as age, sex, years of education, clinical site, BMI and season. Additionally, we employed four machine learning classifiers – Logistic Regression, Decision Tree, Random Forest, and XGBoost – using a nested cross-validation approach to assess the discriminatory capabilities of the RMTs.ResultsThe ANCOVA results indicated significant differences between healthy and AD subjects regarding reduced physical activity, less REM sleep, altered gait patterns, and decreased cognitive functioning. The machine-learning-based analysis demonstrated that RMT-based models could identify subjects in the prodromal stage with an Area Under the ROC Curve of 73.0 %. In addition, our findings show that the Amsterdam iADL questionnaire has high discriminatory abilities.ConclusionsRMTs show promise in AD detection already in the prodromal stage. Using them could allow for earlier detection and intervention, thereby improving patients’ quality of life. Furthermore, the Amsterdam iADL questionnaire holds high potential when employed remotely.

Development and validation of an EHR-based risk prediction model for geriatric patients undergoing urgent and emergency surgery

BackgroundClinical determination of patients at high risk of poor surgical outcomes is complex and may be supported by clinical tools to summarize the patient’s own personalized electronic health record (EHR) history and vitals data through predictive risk models. Since prior models were not readily available for EHR-integration, our objective was to develop and validate a risk stratification tool, named the Assessment of Geriatric Emergency Surgery (AGES) score, predicting risk of 30-day major postoperative complications in geriatric patients under consideration for urgent and emergency surgery using pre-surgical existing electronic health record (EHR) data.MethodsPatients 65-years and older undergoing urgent or emergency non-cardiac surgery within 21 hospitals 2017–2021 were used to develop the model (randomly split: 80% training, 20% test). The primary outcome was a 30-day composite outcome including several postoperative major complications and mortality; secondary outcomes included common individual complications included in the primary composite outcome (sepsis, progressive renal insufficiency or renal failure, and mortality). Patients’ EHR-based clinical history, vital signs, labs, and demographics were included in logistic regression, LASSO, decision tree, Random Forest, and XGBoost models. Area under the receiver operating characteristics curve (AUCROC) was used to compare model performance.ResultsOverall, 66,262 patients (median [IQR] age 78 [(70.9–84.0], female 53.9%, White race 68.5%) received urgent or emergency non-cardiac surgery (25.7% orthopedic cases, 21.9% general surgery cases). AUCROC ranged from 0.655 (Decision Tree) – 0.804 (XGBoost) for the primary composite outcome. XGBoost AUCROC was 0.823, 0.781, and 0.839 in predicting outcomes of sepsis, progressive renal insufficiency or renal failure, and mortality, respectively.ConclusionsWe developed a model to accurately predict major postoperative complications in geriatric patients undergoing urgent or emergency surgery using the patient’s own existing EHR data. EHR implementation of this model could efficiently support clinicians’ surgical risk assessment and perioperative decision-making discussions in this vulnerable patient population.

IoT Network Security Anomaly Detection and Classification using Deep Learning

The Internet of Things (IoT) is an expanding network of interconnected devices exposed to growing cyber security threats. Integrating AI-powered solutions presents a promising avenue for enhancing anomaly detection and classification. This study delves into developing a comprehensive methodology leveraging machine learning and deep learning techniques. Utilizing the BoTNeTIoT-L01 dataset, meticulously curated from IoT devices, the research focuses on data gathering, preprocessing, and exploratory data analysis to unearth underlying patterns and anomalies within network traffic data. Subsequently, a suite of machine learning models, including Logistic Regression, LightGBM (Light Gradient-Boosting Machine), and Decision Tree, along with a deep learning model optimized with the Adam optimizer, is employed to detect and classify anomalies effectively. The comparative analysis underscores the superior performance of advanced models such as LightGBM and Decision Tree, showcasing their efficacy in accurately identifying security threats within IoT environments. The study also addresses pertinent technical challenges, ethical considerations, and future directions, emphasizing the imperative for responsible deployment and ongoing innovation in AI-powered IoT security solutions.

Kidney Disease Prediction

Chronic Kidney Disease (CKD) is a growing public health concern, and early detection is critical for effective intervention. This study compares various machine learning models—Logistic Regression, Decision Trees, Random Forests, Support Vector Machines (SVM), and Neural Networks—to predict CKD. Results show that ensemble methods, like Random Forests, outperform individual models in terms of accuracy and robustness due to their ability to reduce overfitting. However, simpler models like Logistic Regression and Decision Trees offer greater interpretability, which is essential in clinical settings. The study underscores the trade-off between model complexity and transparency, suggesting future research should focus on integrating predictive models with Electronic Health Records (EHRs) to facilitate real-time monitoring and improve patient care. KeyWords: Chronic Kidney Disease, Machine Learning, Random Forest, Logistic Regression, K-Nearest Neighbors, Early Detection.

Analysis of Artificial Intelligence Methods in Classifying Heart Attack Risk: Black-Box Models vs. Glass-Box Models

Artificial Intelligence (AI) is becoming more and more involved in human life day by day. Healthcare is one of the areas where AI is widely used, such as in the diagnosis prediction, and/or classification of diseases. Techniques such as machine learning provide high-accuracy results, but many algorithms have black-box structures, where the reasoning behind the predictions is not known. Explainable AI emerges to address this by providing explanations for complex models. While interpretable ("glass-box") models are desirable, they may have lower accuracy than complex ("black-box") models. Finding the right balance is crucial, especially in critical areas such as healthcare. It is also important to provide individual explanations for the predictions. This study uses patient data to explore a model to predict heart attack risk. Therefore, we compare glass-box models (logistic regression, naive Bayes, decision tree, and explainable boosting) with black-box models (random forest, support vector machine, multi-layer perceptron, gradient boosting, and stochastic gradient boosting). The results show that explainable boosting achieves the highest accuracy. To delve into individual explanations on a patient basis, the explainable boosting algorithm is compared with the random forest algorithm, which gives the best results among the black-box models. Here, LIME and SHAP are used to provide interpretability of random forests. As a result, it is concluded that the random forest algorithm has differences in the importance weights of the variables compared to the explainable boosting algorithm. Both results provide valuable tools for healthcare stakeholders to choose the most appropriate model.

Predicting the churn patterns of monetizers and non-monetizers: exploring the influence of behavioral variability in churn prediction

PurposeAlthough prior research has employed various variables to predict player churn, the dynamic evolution of the behavioral patterns of players has received limited attention. In this study, churn prediction models are developed by incorporating the progress level, in-game purchase, social interaction, behavioral pattern and behavioral variability (BV) of players in social casino games (SCGs). The study distinguishes churn prediction between two player groups: monetizers and non-monetizers.Design/methodology/approachThis study employs three machine learning techniques—logistic regression, decision trees and random forests—using real-world player data from an SCG company to construct churn prediction models. Two experiments were conducted. In Experiment 1, BV was combined with four other variable categories to effectively predict churn behaviors across all players (n = 52,246). In Experiment 2, churn prediction models were developed separately for monetizers (n = 16,628) and non-monetizers (n = 35,618).FindingsThe findings from Experiment 1 indicate that incorporating BV significantly improves the overall performance of churn prediction models. Experiment 2 demonstrates that churn prediction models achieve better performance and predictive accuracy for monetizers and non-monetizers when BV is calculated over the 3-day to 7-day and 7-day to 14-day windows, respectively.Originality/valueThis study introduces BV as a novel variable category for churn prediction, emphasizing within-person variability and demonstrating its effectiveness in enhancing model performance. Churn prediction models were independently constructed for monetizers and non-monetizers, utilizing different time windows for variable extraction. This approach improves predictive performance and highlights key differences in critical variables influencing churn across the two player groups. The findings provide valuable insights into churn management strategies tailored for monetizers and non-monetizers.

Diabetes Prediction Through Linkage of Causal Discovery and Inference Model with Machine Learning Models.

Background/Objectives: Diabetes is a dangerous disease that is accompanied by various complications, including cardiovascular disease. As the global diabetes population continues to increase, it is crucial to identify its causes. Therefore, we predicted diabetes using an AI model and quantitatively examined causal relationships using a causal discovery and inference model. Methods: Kaggle's dataset from the National Institute of Diabetes and Digestive and Kidney Diseases was analyzed using logistic regression, deep learning, gradient boosting, and decision trees. Causal discovery techniques, such as LiNGAM, were employed to infer relationships between variables. Results: The study achieved high accuracy across models using logistic regression (84.84%) and deep learning (84.83%). The causal model highlighted factors such as physical activity, difficulty in walking, and heavy drinking as direct contributors to diabetes. Conclusions: By combining AI with causal inference, this study provides both predictive performance and insight into the factors affecting diabetes, paving the way for tailored interventions.

Structural Equation Modelling of Retinopathy of Prematurity Treatment Integrating Both Physical and Clinical Effects.

Background: We sought to develop a structural equation model (SEM) identifying physical and clinical risk factors associated with treatment for retinopathy of prematurity (ROP). Methods: This retrospective, observational, case-control study included 314 infants screened for ROP between April 2004 and July 2024. A bivariate binary logistic regression model, decision tree, and structural equation model (SEM) were employed to develop a more general model for ROP requiring treatment. Results: In the SEM, the factors significantly associated with ROP treatment included the retinal avascular area according to disk diameter (DD) (p < 0.001), weekly vascularisation rate (DD/w) (p < 0.001), and duration of intubation (days) (p < 0.001). In addition, the following significant associations were identified in both the bivariate analysis and the SEM: lower gestational age (p < 0.001) and birth weight (p <0.001) were associated with greater retinal avascular area; low postnatal weight gain (p < 0.027) was associated with a slow rate of retinal vascularisation; sepsis (p < 0.001), ductus arteriosus (p < 0.001), and the need for transfusion (p < 0.001) were associated with longer intubation mechanical ventilation (IMV). Conclusions: Lower gestational age, lower birth weight, sepsis, ductus arteriosus, transfusion, and lower weight gain increase the risk of requiring ROP treatment. In the SEM, this association is represented through three intermediate physical endogenous variables, namely, the greater temporal avascular area of the retina, the lower postnatal vascularisation rate, and the greater duration of IMV.

Research on Titanic Survival Prediction Based on Machine Learning Method

On April 15, 1912, the British luxury passenger ship Titanic sank on its maiden voyage from Southampton to New York because of a collision with an iceberg, resulting in the death of 1502 out of 2224 passengers and crew. This article gains insight into the factors that influence the survival rate of passengers on the Titanic and establish a model of hard voting consisting of logistic regression, random forest and decision tree to predict what sort of people are more likely to survive in this catastrophe. The process involves dealing with the missing values, creating new variables by feature engineering and fitting the model to the dataset. The overall model performs well accuracy 87.64%. By applying to the navigation field, more data can be collected and more precise predictions can be made. The results can also help individuals to predict risk factors and try to decrease them as much as possible while robustness and stability of the model still need to be refined.

Fetal Health Classification using LightGBM with Grid Search Based Hyper Parameter Tuning

Background: Fetal health monitoring throughout pregnancy is challenging and complex. Complications in the fetal health not identified at the right time lead to mortality of the fetus as well the pregnant women. Hence, obstetricians check the fetal health state by monitoring the fetal heart rate (FHR). Cardiotocography (CTG) is a technique used by obstetricians to access the physical well-being of fetal during pregnancy. It provides information on the fetal heart rate and uterine respiration, which can assist in determining whether the fetus is normal or suspect or pathology. CTG data has typically been evaluated using machine learning (ML) algorithms in predicting the wellness of the fetal and speeding up the detection process. Methods: In this work, we developed LightGBM with a Grid search-based hyperparameter tuning model to predict fetal health classification. The classification results are analysed quantitatively using the performance measures, namely, precision, Recall, F1-Score, and Accuracy Comparisons were made between different classification models like Logistic Regression, Decision Tree, Random Forest, k-nearest neighbors, Bagging, ADA boosting, XG boosting, and LightGBM, which were trained with the CTG Dataset obtained by the patented fetal monitoring system of 2,216 data points from pregnant women in their third trimester available in the Kaggle dataset. The dataset contains three classes: normal, suspect, and pathology. Our proposed model will give better results in predicting fetal health classification. Results: In this paper, the performance of the proposed algorithm LightGBM is compared and experimented with various Machine learning Techniques namely LR, DT, RF, KNN, Boosting, Ada boosting, and XG Boost and the classification accuracy of the respective algorithms are 84%, 94%, 93%, 88%, 94%, 89%, 96%.The LightGBM achieved a performance of 97% and outperforms the former models. Conclusion: The LightGBM-based fetal health classification has been presented. Ensemble models were applied to the FHR dataset and presented the hybrid algorithm, namely Light GBM, and its application to fetal health classification. LightGBM has advantages that include fast training, improved performance, scale-up capabilities, and lesser memory usage than other ensemble models. The proposed model is more consistent and superior to other considered machine learning models and is suitable for the classification of fetal health based on FHR data. Finally, the outcomes of the multiple methods are compared using the same training and test data in order to verify the efficiency of LightGBM. The model can be further enhanced by making it hybrid by combining the advantages of different models and optimization techniques.

A strategy to enhance decisions in logistics and transportation companies

With the sanitizing robot project for classrooms, where nowadays a problem faced by the population is the health crisis generated since 2019 called COVID-19, considering that in this year 2024, with a new wave, measures in face-to-face classes only involve protocols for applying hand sanitizer and wearing masks, without taking into account the time spent inside the classrooms, which is almost the entire day Therefore, if a sanitizing robot controlled by Arduino through a mobile application is developed to disinfect the classrooms of the mechatronics engineering program in Building F at the Technological Institute of Higher Studies in Jilotepec, it can be controlled via a mobile phone using Bluetooth and sanitizes using a 360° rotating nozzle to reach all areas, while also having an obstacle detection sensor to prevent collision

EXPLORATION OF STUDENTS INTERESTS IN MBKM AT RIAU UNIVERSITY USING A MACHINE LEARNING APPROACH

This study aims to analyze the factors that have a significant influence on the interest of Riau University students in the Merdeka Belajar Kampus Merdeka (MBKM) program using a machine learning approach. MBKM is an innovation initiated by the Ministry of Education and Culture with the aim of improving student competence through its various programs. The Riau University as one of the universities supports this program by providing opportunities for its students to participate in various activities provided in the MBKM program. This study will specifically use a machine learning approach by utilizing several methods to analyze significant factors that have not been analyzed in depth by previous studies. The methods used in this analysis are logistic regression, decision trees, random forests, and naive bayes by utilizing secondary data on the level of interest of Riau University students to participate in the MBKM program in 2023. The variables used in this study include gender, generation, faculty, knowledge, self-confidence, feeling benefits, family support, friend support, lecturer support, self-ability, and facilities as independent variables and MBKM interest as a dependent variable. The results of the analysis of several methods show that the logistic regression method provides the best performance in modeling with an accuracy level of 95%. Variables that have a significant influence on students' interest in the MBKM program have also been successfully identified. The variables that have a significant effect are self-ability and family support. The development strategy of MBKM at the University of Riau can be optimized by paying attention to and focusing on these variables. The optimization of this strategy aims to make the implementation of the program more effective and efficient. Supportive policies such as workshops for the development of students' soft skills can be one of the strategic steps to improve students' abilities to the maximum

A machine learning tool for early identification of celiac disease autoimmunity

Identifying which patients should undergo serologic screening for celiac disease (CD) may help diagnose patients who otherwise often experience diagnostic delays or remain undiagnosed. Using anonymized outpatient data from the electronic medical records of Maccabi Healthcare Services, we developed and evaluated five machine learning models to classify patients as at-risk for CD autoimmunity prior to first documented diagnosis or positive serum tissue transglutaminase (tTG-IgA). A train set of highly seropositive (tTG-IgA > 10X ULN) cases (n = 677) with likely CD and controls (n = 176,293) with no evidence of CD autoimmunity was used for model development. Input features included demographic information and commonly available laboratory results. The models were then evaluated for discriminative ability as measured by AUC on a distinct set of highly seropositive cases (n = 153) and controls (n = 41,087). The highest performing model was XGBoost (AUC = 0.86), followed by logistic regression (AUC = 0.85), random forest (AUC = 0.83), multilayer perceptron (AUC = 0.80) and decision tree (AUC = 0.77). Contributing features for the XGBoost model for classifying a patient as at-risk for undiagnosed CD autoimmunity included signs of anemia, transaminitis and decreased high-density lipoprotein. This model’s ability to distinguish cases of incident CD autoimmunity from controls shows promise as a potential clinical tool to identify patients with increased risk of having undiagnosed celiac disease in the community, for serologic screening.

Association between serum hypertriglyceridemia and hematological indices: data mining approaches

BackgroundHigh triglyceride (TG) affects and is affected of other hematological factors. The determination of serum fasted triglycerides concentrations, as part of a lipid profile, is crucial key point in hematological factors and significantly affect various systemic diseases. This study was carried out to assess the potential relation between the concentration of TG and hematological factors.MethodOur sample size was 9704 participants beginning in 2007 and ending in 2020 aged between 35 and 65 years, sourced from the MASHAD cohort (northeastern Iran). Machine learning methodologies, specifically logistic regression, decision tree, and random forest algorithms, were utilized for data analysis in the investigation of individuals with normal and high TG levels.ResultsThe highest Gini score belongs to RLR (Red cell distribution width/Lymphocyte) (236.10), RPR (Red cell distribution width/Platelets) (215.78), and PHR (Platelets/high-density lipoprotein) (273.66). We also found that factors such as age are statistically associated with the level of TG in women probably due to the drop in menopausal estrogen. RF model showed to have higher accuracy in predicting the TG level in both males and females.ConclusionOur model assessed the association between serum TG with several hematological factors like RLR, RPR, and PHR. Other hematological factors also have been reported to be related to the TG level. As these results give us new insights into the association of TG on various hematological factors and their possible interactions with each other. future studies are needed to provide sufficient data for the mechanism and the pathophysiology of the findings.

Machine Learning for Health Insurance Prediction in Nigeria

Health insurance coverage remains critical to healthcare accessibility, particularly in developing nations like Nigeria. This paper focused on predicting the likelihood of medical insurance coverage among individuals in Nigeria by employing four prominent Machine learning techniques: Logistic Regression, Random Forest, Decision Tree, and Support Vector Machine classifiers. The dataset utilized for analysis comprises demographic information, socioeconomic factors, and health-related variables collected from a diverse sample across Nigeria. Four models are trained and evaluated: Logistic Regression widely accepted for its simplicity and interpretability. Random Forest is a robust ensemble learning algorithm capable of capturing complex relationships within the data. The decision Tree model is simple to understand and visualize and the Support Vector Machine model is known for producing a very good classification. Furthermore, the performance metrics uutilized to rate the predictive capabilities of the models are Accuracy, Precision, Sensitivity, F Score, and area under the Receiver Operating Characteristic (AUC &amp; ROC Curve). Additionally, a features importance analysis is conducted for the identification of the dominant factors contributing to the prediction of the spread of medical insurance in Nigeria. The outcome of this paper gives insights in the efficiency of each machine learning models used to forecast medical insurance coverage, and identifying key determinants influencing insurance coverage can assist policymakers and healthcare stakeholders in devising targeted strategies to improve healthcare access and affordability for the Nigerian people.