Variable Ranking Research Articles

Machine learning unravels controls on river water temperature regime dynamics

Water temperature is vitally important to the health of rivers and streams, influencing the integrity of ecosystems, aquatic biogeochemistry, and the habitability of waterways for a variety of species. While climate is often regarded as the primary driver of stream temperature regimes, other factors - including hydrology, watershed characteristics, and human impacts - add substantial complexity to the variability of water temperatures. However, it remains challenging to disentangle the influence of these drivers through time and across rivers spanning diverse settings. To quantify the underlying controls on river thermal regimes, we applied conditional inference random forest models to predict maximum monthly stream temperatures and thermal sensitivities, averaged across a 4-year period (2016 to 2020), at 410 watersheds spanning the conterminous United States. Maximum stream temperatures were selected given their ecological relevance, while thermal sensitivity, which measures the relationship between air and water temperatures, was used to assess the responsiveness of stream temperatures to climate forcings. We interpreted these random forest models using variable importance rankings, describing seasonal and spatial variability in the dominant controls on water temperatures. Although our empirical results confirm that climate is indeed a primary control on temperature magnitude, our models highlight the diversity in drivers of water temperature variability across seasons, hydrologic regions, and between metrics. By combining random forest models with process-based understanding of stream thermal regimes, we provide new insights on the dynamic controls of water temperature variability across broad geographical domains, informing region- and season-specific controls for tailored thermal watershed management and guiding the framing of future water temperature modeling.

Identifying potential significant factors impacting zero-inflated proportion data.

Classical supervised methods like linear regression and decision trees are not completely adapted for identifying impacting factors on a response variable corresponding to zero-inflated proportion data (ZIPD) that are dependent, continuous and bounded. In this article we propose a within-block permutation-based methodology to identify factors (discrete or continuous) that are significantly correlated with ZIPD, we propose a performance indicator quantifying the percentage of correlation explained by the subset of significant factors, and we show how to predict the ranks of the response variables conditionally on the observation of these factors. The methodology is illustrated on simulated data and on two real data sets dealing with epidemiology. In the first data set, ZIPD correspond to probabilities of transmission of Influenza between horses. In the second data set, ZIPD correspond to probabilities that geographic entities (eg, states and countries) have the same COVID-19 mortality dynamics.

Association between cognitive impairment and main metals among oldest old aged 80 years and over in China

Objective: To identify the main metals involved in cognitive impairment in the Chinese oldest old, and explore the association between these metal exposures and cognitive impairment. Methods: A cross-sectional study was conducted on 1 568 participants aged 80 years and older from Healthy Aging and Biomarkers Cohort Study (2017 to 2018). Fasting venous blood was collected to measure the levels of nine metals (selenium, lead, cadmium, arsenic, antimony, chromium, manganese, mercury, and nickel). The cognitive function of these participants was evaluated by using the Chinese version of the Mini-Mental State Examination (CMMSE). The random forest (RF) was applied to independently identify the main metals that affected cognitive impairment. The multivariate logistic regression model and restricted cubic splines (RCS) model were used to further verify the association of the main metals with cognitive impairment. Results: The age of 1 568 study subjects was (91.8±7.6) years old, including 912 females (58.2%) and 465 individuals (29.7%) with cognitive function impairment. Based on the RF model (the out-of-bag error rate was 22.9%), the importance ranking of variables was conducted and the feature screening of five times ten-fold cross-validation was carried out. It was found that selenium was the metal that affected cognitive function impairment, and the other eight metals were not included in the model. After adjusting for covariates, the multivariate logistic regression model showed that with every increase of 10 μg/L of blood selenium levels, the risk of cognitive impairment decreased (OR=0.921, 95%CI: 0.889-0.954). Compared with the lowest quartile(Q1) of blood selenium, the ORs (95%CI) of Q3 and Q4 blood selenium were 0.452 (0.304-0.669) and 0.419 (0.281-0.622) respectively. The RCS showed a linear dose-response relationship between blood selenium and cognitive impairment (Pnonlinear>0.05). Conclusion: Blood selenium is negatively associated with cognitive impairment in the Chinese oldest old.

Tracing the geographical origin of tobacco at two spatial scales by stable isotope and element analyses with chemometrics.

Tobacco is a widely cultivated cash crop, but it is often smuggled and sold illegally. Unfortunately, there is currently no way to verify the origin of tobacco in China. In an effort to address this issue, we conducted a study using stable isotopes and elements from 176 tobacco samples at both provincial and municipal scales. Our findings revealed significant differences in δ13C, K, Cs, and 208/206Pb at the provincial-level, and Sr, Se, and Pb at the municipal level. We created a heat map at the municipal level, which showed a similar cluster classification to geographic grouping and provided an initial assessment of tobacco origins. Using OPLS-DA modeling, we achieved a 98.3% accuracy rate for the provincial scale and 97.6% for the municipal scale. It is worth noting that the importance of rankings of variables varied depending on the spatial scale of the evaluation. This study offers the first traceability fingerprint dataset of tobacco and has the potential to combat mislabeling and fraudulent conduct by identifying the geographical origin of tobacco.

Building risk mitigation strategies for circularity adoption in Indian textile supply chains.

Textile industries are among the most polluting and demand urgent management measures to mitigate their negative environmental impact. Thus, it is imperative to incorporate the textile industry into the circular economy and to foster sustainable practices. This study aims to establish a comprehensive, compliant decision framework to analyse risk mitigation strategies for circular supply chain (CSC) adoption in India's textile industries. The Situations Actors Processes and Learnings Actions Performances (SAP-LAP) technique analyses the problem. However, interpreting the interacting associations between the SAP-LAP model-based variables is somewhat lacking in this procedure, which might skew the decision-making process. As a result, in this study, the SAP-LAP method is accompanied by a novel ranking technique, namely, the Interpretive Ranking Process (IRP), which reduces decision-making issues in the SAP-LAP method and aids in evaluating the model by determining the ranks of variables; furthermore, the study also offers causal relationships among the various risks and risk factors and various identified risk-mitigation actions by constructing Bayesian Networks (BN) based on conditional probabilities. The study's originality represents the findings using an instinctive and interpretative choice approach to address significant concerns in risk perception and mitigation techniques for CSC adoption in the Indian textile industries. The suggested SAP-LAP and the IRP-based model would assist firms in addressing risk mitigation techniques for CSC adoption concerns by providing a hierarchy of the various risks and mitigation strategies to cope with. The simultaneously proposed BN model will help visualise the conditional dependency of risks and factors with proposed mitigating actions.

A contingency approach for time-cost trade-off in construction projects based on machine learning techniques

PurposeTime-cost trade-off is normal conduct in construction projects when projects are expectedly late for delivery. Existing research on time-cost trade-off strategic management mostly focused on the technical calculation towards the optimal combination of activities to be accelerated, while the managerial aspects are mostly neglected. This paper aims to understand the managerial efforts necessary to prepare construction projects ready for an upcoming trade-off implementation.Design/methodology/approachA preliminary list of critical factors was first identified from the literature and verified by a Delphi survey. Quantitative data was then collected by a questionnaire survey to first shortlist the preliminary factors and quantify the predictive model with different machine learning algorithms, i.e. k-nearest neighbours (kNN), radial basis function (RBF), multiplayer perceptron (MLP), multinomial logistic regression (MLR), naïve Bayes classifier (NBC) and Bayesian belief networks (BBNs).FindingsThe model's independent variable importance ranking revealed that the top challenges faced were the realism of contractual obligation, contractor planning and control and client management and monitoring. Among the tested machine learning algorithms, multilayer perceptron was demonstrated to be the most suitable in this case. This model accuracy reached 96.5% with the training dataset and 95.6% with an independent test dataset and could be used as the contingency approach for time-cost trade-offs.Originality/valueThe identified factor list contributed to the theoretical explanation of the failed implementation in general and practical managerial improvement to better avoid such failure. In addition, the established predictive model provided an ad-hoc early warning and diagnostic tool to better ensure time-cost implementation success.

The impact of Covid-19 on the Korean and US labour markets

ABSTRACT This study compares labour market experiences in South Korea and the US at the outbreak of the pandemic and then again in November 2020. We found that the pandemic had the most considerable effect on the not-at-work rate in South Korea and the unemployment rate in the US. We computed concentration indices to measure inequality in the labour markets using education as a socioeconomic ranking variable. Applying a Recentered Influence Function (RIF) regression, we found that unemployment was more concentrated among less-educated workers in South Korea. Still, the not-at-work rate was more concentrated among highly educated workers. While the ability to work from home played an important role in explaining these inequalities, by November 2020, the Korean labour market showed minimal disparities. In general, US workers with lower education levels experienced higher unemployment and not-at-work rates. The capability to work remotely considerably reduced inequality in April, but it did not in November.

(201) Young Adult Males Report Reluctance in Disclosing Sexual Health Concerns to Physicians: A Cross-Sectional Study

Abstract Introduction The reported prevalence of sexual dysfunction in young adult males has recently increased, but estimates are highly variable, ranging from 1% to 30%. Sexual dysfunction can have organic, psychogenic, and multidimensional causes, and sexual concerns warrant medical assessment. However, both clinical management and epidemiologic research depend upon the willingness of patients to disclose symptoms. Despite this issue, there is limited research on the comfort level of young adult males in discussing sexual health or sexual dysfunction with physicians. Objective To assess the comfort level of young adult males in discussing their sexual health, including sexual side effects caused by medications, with doctors. Methods A cross-sectional, paid survey was designed in collaboration with our institution’s biostatistics core using Qualtrics XM. Biological males ages 18 to 35 were recruited using social media. The survey included demographic characteristics and contained several 5-point Likert scale questions to query perceptions of personal health. Additionally, respondents ranked the top three of seven subjects that they felt most difficult to talk about with their doctor (mental health, physical health, school/job achievement, maintaining body weight, ability to orgasm, libido, and erectile function). Participants also ranked the three items that were most important to them according to their current personal values. These ranked variables were subsequently transformed to calculate an “impact score” for analysis (3 points = ranked 1st; 2 points = ranked 2nd; 1 point = ranked 3rd; 0 points = unranked). Results 505 out of 665 electronically delivered surveys were returned (76% response rate), and 486 had complete and evaluable responses. Age was normally distributed with a mean of 27.4 years (SD: 3.6). Participants were 40% non-white, 21% LGBTQ+, and 89% sexually active. 37% reported a history of major depressive disorder (MDD) and 34% reported a history of generalized anxiety disorder (GAD). 51% of respondents agreed or strongly agreed with the statement, “I am embarrassed to talk about sexual problems with my doctor.” As illustrated by Figure 1A, respondents ranked all sexual health topics as uniformly and significantly more difficult to discuss with their physicians compared to all other aspects of personal health assessed (p=0.0000-0.02). As displayed in Figure 1B, respondents indicated that sexual health subjects were of lower importance relative to the other areas of personal health. Grossly, there were no significant correlations between impact scores and age. Conclusions Our results suggest that young adult males display high rates of embarrassment and reluctance in sharing sexual concerns with physicians. Though young adult males place less value on sexual health, it is closely interconnected with mental and physical well-being. Our findings reinforce the crucial need for physicians to actively combat stigma around sexual health conversations, normalize these discussions with their young adult male patients, and directly inquire about sexual concerns in patients with medical conditions or medications with known sexual side effects. Disclosure Any of the authors act as a consultant, employee or shareholder of an industry for: Dadi, Teleflex, Lina.

The $$\kappa $$-nullity of Riemannian manifolds and their splitting tensors

We consider Riemannian n-manifolds M with nontrivial $$\kappa $$ -nullity “distribution” of the curvature tensor R, namely, the variable rank distribution of tangent subspaces to M where R coincides with the curvature tensor of a space of constant curvature $$\kappa $$ ( $$\kappa \in \mathbb {R}$$ ) is nontrivial. We obtain classification theorems under diferent additional assumptions, in terms of low nullity/conullity, controlled scalar curvature or existence of quotients of finite volume. We prove new results, but also revisit previous ones.

A data‐analytic framework to monitor product density of four‐effect falling‐film evaporator for skimmed milk production

AbstractThe implementation of Industry 4.0 needs the convergence in operational technology and information technology (OT–IT) for improving the flexibility to access real‐time data and KPIs. In India, a small‐scale automation industry is full of edge devices and produces massive amount of data due to which an infrastructure may overwhelm. These industries face challenges in convergence of IT–OT like scalability. This work developed a data‐driven framework, which is an additional infrastructure between the IT–OT platforms. The framework consists of data cleaning which helps to remove the anomaly from the sensor data. The uncertainty analysis ensures the correctness of the data. Variable ranking and correlation matrix help to identify the KPIs of the process. All these stages are helpful to select the data and push this selected data to cloud for developing predictive or diagnostic models. This framework is deployed for the performance prediction of four‐effect falling‐film evaporator used to produce skimmed milk. The outlet skimmed milk density is the key performance indicator for the present case study and it is monitored here. Both linear and neural network‐based nonlinear models are used in the framework. It is noticed that linear regression model is capable of predicting the plant product density of skimmed milk with good accuracy. However, it is not capable to identify the faults in the key parameters like inlet milk flow rate. These faults can be better identified using k‐means clustering at the initial stage of the framework.Practical applicationsThis article titled “A data‐analytic framework to monitor product density of four‐effect falling‐film evaporator for skimmed milk production” by Nivedita Wagh and Sudhir D. Agashe contains the results obtained for data‐driven framework for performance prediction and fault diagnosis for four‐effect falling‐film evaporator used for production of skimmed milk. A data‐driven framework which includes uncertainty analysis, variable ranking, various linear and neural network‐based nonlinear regression models for performance prediction, and fault detection of an evaporator is proposed. The results are presented in terms of density of the skimmed milk as a performance parameter. The objective of the work is to provide the possibility of IT–OT convergence so as to solve the scalability issue in the IT–OT convergence. The authors are able to bring out a data‐driven framework for fault identification using k‐means clustering using four‐effect falling‐film evaporator used for production of skimmed milk as a case study. We hope the results will be useful to data managers for understanding the KPIs statistics and take appropriate decision in time for the overall growth of organization.

Machine-Learning Classification Models to Predict Liver Cancer with Explainable AI to Discover Associated Genes

Hepatocellular carcinoma (HCC) is the primary liver cancer that occurs the most frequently. The risk of developing HCC is highest in those with chronic liver diseases, such as cirrhosis brought on by hepatitis B or C infection and the most common type of liver cancer. Knowledge-based interpretations are essential for understanding the HCC microarray dataset due to its nature, which includes high dimensions and hidden biological information in genes. When analyzing gene expression data with many genes and few samples, the main problem is to separate disease-related information from a vast quantity of redundant gene expression data and their noise. Clinicians are interested in identifying the specific genes responsible for HCC in individual patients. These responsible genes may differ between patients, leading to variability in gene selection. Moreover, ML approaches, such as classification algorithms, are similar to black boxes, and it is important to interpret the ML model outcomes. In this paper, we use a reliable pipeline to determine important genes for discovering HCC from microarray analysis. We eliminate redundant and unnecessary genes through gene selection using principal component analysis (PCA). Moreover, we detect responsible genes with the random forest algorithm through variable importance ranking calculated from the Gini index. Classification algorithms, such as random forest (RF), naïve Bayes classifier (NBC), logistic regression, and k-nearest neighbor (kNN) are used to classify HCC from responsible genes. However, classification algorithms produce outcomes based on selected genes for a large group of patients rather than for specific patients. Thus, we apply the local interpretable model-agnostic explanations (LIME) method to uncover the AI-generated forecasts as well as recommendations for patient-specific responsible genes. Moreover, we show our pathway analysis and a dendrogram of the pathway through hierarchical clustering of the responsible genes. There are 16 responsible genes found using the Gini index, and CCT3 and KPNA2 show the highest mean decrease in Gini values. Among four classification algorithms, random forest showed 96.53% accuracy with a precision of 97.30%. Five-fold cross-validation was used in order to collect multiple estimates and assess the variability for the RF model with a mean ROC of 0.95±0.2. LIME outcomes were interpreted for two random patients with positive and negative effects. Therefore, we identified 16 responsible genes that can be used to improve HCC diagnosis or treatment. The proposed framework using machine-learning-classification algorithms with the LIME method can be applied to find responsible genes to diagnose and treat HCC patients.

A meta-analysis of dietary metabolizable amino acids and energy supply on nitrogen retention and nitrogen utilization efficiency in beef cattle

Increasing nitrogen utilization efficiency (NUE) can reduce environmental impact and increase economic benefits to maintain a sustainable development of the beef cattle industry. The objectives of this study were to systemically and quantitatively analyze the effects of dietary metabolizable amino acids and energy supply on nitrogen retention and NUE in beef cattle and to develop prediction models using a meta-regression and a machine learning approach. After literature screening and data collection, a dataset was assembled, which finally included 63 peer-reviewed studies published between 1980 and 2022 with 248 treatments. Following bivariate correlation analyses, univariate and multiple linear regression analyses were conducted to analyze effects of body weight (BW) and intake of metabolizable energy (MEI), crude protein (CPI), metabolizable protein (MPI), and 10 metabolizable essential amino acids (MAAI) on retained protein and NUE. The mixed model procedure with study as random effect was used to remove variations across studies. The random forest regression model was used to identify the importance ranking of predictor variables and to compare prediction adequacy with the multiple linear regression models. The stability of parameter estimates and the validity of the obtained model were evaluated using the Monte Carlo cross-evaluation approach. Both retained protein and NUE were negatively quadratically correlated with BW, MEI, CPI, MPI, and MAAI, suggesting that the marginal return of retained protein and NUE are decreased with increased dietary energy and protein supply. The multiple linear regression analyses indicated that retained protein and NUE were negatively correlated with BW, while positively correlated with MEI and metabolizable Arg intake. However, there was a negative correlation between NUE and metabolizable tryptophan intake. Compared with the multiple regression models, the random forest models had greater RMSPE and lesser CCC, suggesting that the prediction accuracy and precision of the random forest models were less than the multiple regression models. In conclusion, the conversion efficiency of MAAI to net protein is not constant, and the developed response surface models can be a reliable approach to estimate AA requirements and optimize NUE in beef cattle.

Maternal Variability of Amplitudes of Frequency Fluctuations Is Related to the Progressive Self-Other Transposition Group Intervention in Autistic Children.

The self-to-other model of empathy (SOME) states that a key reason for the empathic deficiency in autistic individuals is the imbalance of the self-other switch. The existing interventions of theory of mind contain training of self-other transposition ability but combined with other cognitive trainings. The self-other distinction brain areas of autistic individuals have been revealed, but the brain areas of the self-other transposition ability and its intervention have not been investigated. There are normalized amplitudes of low-frequency fluctuations (mALFFs) within 0.01-0.1 Hz and many normalized amplitudes of frequency fluctuations (mAFFs) within 0-0.01, 0.01-0.05, 0.05-0.1, 0.1-0.15, 0.15-0.2, and 0.2-0.25 Hz. Therefore, the current study established a progressive self-other transposition group intervention to specifically and systematically improve autistic children's self-other transposition abilities. The transposition test with a three mountains test, an unexpected location test, and a deception test was used to directly measure autistic children's transposition abilities. The Interpersonal Responsiveness Index Empathy Questionnaire with perspective-taking and fantasy subscales (IRI-T) was used to indirectly measure autistic children's transposition abilities. The Autism Treatment Evaluation Checklist (ATEC) was used to measure autistic children's autism symptoms. The experiment was designed with two (intervention: experimental group vs. control group) independent variables and two (test time: pretest vs. posttest or tracking test) × three (test: transposition test vs. IRI-T test vs. ATEC test) dependent variables. Furthermore, it used eyes-closed resting-state functional magnetic resonance imaging to investigate and compare the relevant maternal mALFFs and average energy rank and energy rank variability of mAFFs of autistic children's transposition abilities, autism symptoms, and intervention effects. The results showed the following: (1) There were many improvements (pretest vs. posttest or tracking test) greater than chance 0 in the experimental group, such as the three mountains, lie, transposition, PT, IRI-T, PT tracking, cognition, behavior, ATEC, language tracking, cognition tracking, behavior tracking, and ATEC tracking improvements. However, there was no improvement greater than chance 0 in the control group. (2) The maternal mALFFs and maternal average energy rank and energy rank variability of mAFFs could predict the autistic children's transposition abilities, autism symptoms, and intervention effects with some overlap and some difference in maternal self-other distinction, sensorimotor, visual, facial expression recognition, language, memory and emotion, and self-consciousness networks. These results indicated that the progressive self-other transposition group intervention successfully improved autistic children's transposition abilities and reduced their autism symptoms; the intervention effects could be applied to daily life and last up to a month. The maternal mALFFs, average energy rank, and energy rank variability of mAFFs were three effective neural indictors of autistic children's transposition abilities, autism symptoms, and intervention effects, and the average energy rank and energy rank variability of mAFFs were two new neural indictors established in the current study. The maternal neural markers of the progressive self-other transposition group intervention effects for autistic children were found in part.

Impact of a Simulation Game (MASS) on the Undergraduate’s Experience and Performance in Triage Training–A Pilot Study

Introduction:The use of simulation games in education has been rising in recent years. Triage is not only a major content among the EMS (emergency medical services) but also a necessary skill for students to improve their first-aid ability. This study aims to investigate a game for education called MASS (movement, assessment, sort, and sending), which attempts to enhance students’ capability of disaster response.Method:A randomized controlled trial was conducted among students who took the course: “Understanding Disaster and Surviving Risk” during the term and volunteered to participate in teaching research from different faculties in Sichuan University. Participants were trained by using the simulation game or the online course before class. The simulation game is MASS, which uses virtual reality techniques to create a realistic 3D tanker explosion scene, and the online course is a Massive Open Online Course created by the teaching team. In the class, questionnaires with subjective and objective multiple choices were carried out after a discussion. With SPSS version 27.0, statistical significance among groups was determined by Mann-Whitney U-test for rank variables, Fisher’s exact test for binary variables.Results:73 students were included in this study, including 45 in the online course group and 28 in the simulation game group. The qualitative experience survey showed the two groups were significantly different in experience of disaster scene challenges, application scenes, and method of triage (P=0.031, 0.007, 0.031, respectively). Students in the simulation game group showed significantly better performance in knowledge acquisition including application scene of triage, key of the expectant’s assessment, and rescue condition for thoracentesis than in the Online game group (P=0.048, 0.020, 0.010, respectively).Conclusion:Simulation games can improve students’ experience and performance in triage training. Due to the unrepeatability of disaster, the study suggests that games can be used to conduct simulation education for disaster medicine.

High-dimensional CFD optimization of a low-flow coefficient S–CO2 centrifugal compressor for enhanced oil recovery systems

The design of low-flow coefficient (∼0.01) centrifugal compressors with supercritical CO2 as working fluid is still a challenge for engineers due to its increased friction losses at the impeller. However, the reinjection pressure required for Enhanced Oil Recovery (EOR) systems is achieved by compression trains with stages of high-Pressure Ratios (PR > 3) which can only be obtained by lowering the flow coefficient of the equipment. The carbon dioxide mitigation, due to the reinjection process, also increases oil productivity and extraction lifetime. A four-staged compression system was considered and the preliminary geometry of its last stage was considered herein after a 1D optimization that decreased the total required power of the system. In order to further increase the systems' performance, a CFD model was developed and submitted to Sensitivity Analysis (SA) and parametric optimization procedure, considering polar angles, meridional profile and vaneless diffuser passage (25 variables). The assessment of the sequential SA using, Morris' screening method Design of Experiment (DoE) and SS-ANOVA for variable ranking and response surface training, has exposed the method's limitation in recognizing interaction between variables since low-quality Response Surfaces (RS) were trained. However, the Incremental Space Filler (ISF) sampling has complemented the sample space screening, guarantying adequate RS at a low computational cost. This indirect optimization strategy that increased the equipment's polytropic efficiency by 1.19%, diminishing total entropy generation by 8.5% can deliver important cost reductions to the operation of EOR compression systems. The ‘entropy-guided’ phenomenology analysis strategy combined with SA results, identified that the narrowing of the vaneless diffuser has extinguished the recirculation present in the original geometry's impeller/diffuser interface region, which was the largest difference in the entropy histogram. Moreover, the enlargement of the impeller's meridional profile has smoothed the fluid flow change of direction (from axial to radial) and displaced the swirl structures that restricted the fluid flow in the main passage.

Income-related inequality in out-of-pocket health-care expenditures under Taiwan's national health insurance system: An international comparable estimation based on A System of Health Accounts

National estimates of out-of-pocket health-care expenditures (OOP-HCEs) that use comparable international guidelines based on A Systems of Health Accounts (SHA) are generally unavailable in Taiwan. International comparable OOP are essential for designing universal health-coverage (UHC) policy. We designed an SHA-based household OOP questionnaire. A nationally representative cross-sectional survey was then conducted from January to August 2022. The final questionnaire was completed by 657 households and 1969 individuals. The total OOPs were divided into expenditures related to curative care (HC.1), rehabilitative care (HC.2), long-term care (HC.3), ancillary services (HC.4), and medical goods (HC.5). National estimates were calculated by accounting for the complex survey design. Variance was estimated through Taylor series linearization. The concentration index was calculated using household income as the ranking variable. We then identified factors contributing to the inequality in OOP distribution by household income. National estimates revealed an OOP of NT$424 billion, which accounted for 29.6% of Taiwan's national health expenditure in 2021. Private health insurance (PHI) reimbursements accounted for 9.0% of the total OOP. The OOPs for curative care and medical goods accounted for 50.1% and 39.0% of the total OOP, respectively. The OOPs after PHI reimbursements were progressive (concentration index = 0.103, P = 0.012). The frequency of medical-care use and the number of medical visits negatively affected progressive OOPs. International comparable OOPs revealed that under the Taiwanese National Health Insurance (NHI), OOPs can still be high. However, the NHI might have caused OOPs to be progressive from the perspective of income but regressive from the perspective of health status. Countries striving for UHC should consider the redistribution effect of public health insurance and possible inequalities in health.

469 Use of data-driven technology to optimise DOACs in morbidly obese patients

Abstract Introduction Despite the advantages of Direct Oral Anticoagulants (DOACs) over older classes of anticoagulants, clinical experience is limited in special populations; data from landmark trials on safety and efficacy are relatively scarce (compared to warfarin). This makes it challenging for clinicians to prescribe the right DOAC at the right dose for such patients (e.g., morbidly obese patients) (1). Insights derived from analysing real-world data have proven to be a vital source of clinical evidence backing the recommendation of medications (2). Therefore, data-driven technologies like machine learning can harness big data in electronic health records (EHRs) to optimise DOAC therapy and improve clinical outcomes. Aim The study aims to accurately predict clinical outcomes in morbidly obese patients, and identify the key variables in the model for optimising the safety and efficacy of Direct Oral Anticoagulant (DOAC) doses. Methods An observational, retrospective cohort study was carried out in partnership with an NHS Trust. Based on eligibility criteria, the dataset of morbidly obese patients on DOACs was extracted from EHRs, pre-processed and analysed considering the access granted. After partitioning the entire dataset into a 70:30 split, the training dataset (70%) was run through selected machine learning (ML) classifiers (Random Forest, decision trees, K-nearest neighbours (KNN), bootstrap aggregation algorithm, gradient boosting classifier, support vector machines, and logistic regression) to rank variables, and derive predictions which were evaluated against the test dataset (30%). A multivariate regression model was used to adjust for confounders and to explore the relationships between DOAC regimens and clinical outcomes. Results We identified 4,275 morbidly obese patients out of n=97,413 records overall. The bootstrap aggregation, decision trees, and random forest classifiers (from the ML algorithms tested) achieved superior prediction accuracies (98.6%, 97.9%, and 98.3%, respectively) for the individual DOAC doses, with excellent values for precision, recall, and F1 scores (performance metrics). The most important characteristics in the model for predicting mortality and stroke were age, treatment days, and length of stay. Among DOACs, apixaban (84%) was the most frequently prescribed DOAC followed by rivaroxaban (15%). Apixaban 2.5 mg (twice daily) received the highest ranking for relevance to mortality, while it raised the mortality risk (OR 1.430, 95% CI: 1.181, 1.1.732, p=0.001). There were mixed results for apixaban 5mg (twice daily), the most widely prescribed dose of apixaban (54%), with significantly reduced risk of mortality (OR 0.751, 95% CI: 0.632, 0.905, p=0.003), but significantly increased risk of stroke events (OR 32.457, 95% CI: 17.083-61.664, p=0.001). Conclusion Given the large sample size—a strength of our study, data-driven technologies were successfully employed in predicting the safety and efficacy of DOACs in morbidly obese patients using the real-world dataset; the key variables in the model for optimising clinical outcomes were identified. However, the limitations in our study, such as reporting errors, selection bias, and confounding bias, were not ruled out. Therefore, confirmatory studies (e.g., external validation with prospective data) are needed to confirm findings and provide a sound basis for universal deployment in clinical settings.

Application of machine learning model based on XGBoost algorithm in early prediction of patients with acute severe pancreatitis

To establish a machine learning model based on extreme gradient boosting (XGBoost) algorithm for early prediction of severe acute pancreatitis (SAP), and explore its predictive efficiency. A retrospective cohort study was conducted. The patients with acute pancreatitis (AP) who admitted to the First Affiliated Hospital of Soochow University, the Second Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University from January 1, 2020 to December 31, 2021 were enrolled. Demography information, etiology, past history, and clinical indicators and imaging data within 48 hours of admission were collected according to the medical record system and image system, and the modified CT severity index (MCTSI), Ranson score, bedside index for severity in acute pancreatitis (BISAP) and acute pancreatitis risk score (SABP) were calculated. The data sets of the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University were randomly divided into training set and validation set according to 8 : 2. Based on XGBoost algorithm, the SAP prediction model was constructed on the basis of hyperparameter adjustment by 5-fold cross validation and loss function. The data set of the Second Affiliated Hospital of Soochow University was served as independent test set. The predictive efficacy of the XGBoost model was evaluated by drawing the receiver operator characteristic curve (ROC curve), and compared it with the traditional AP related severity score; variable importance ranking diagram and Shapley additive explanation (SHAP) diagram were drawn to visually explain the model. A total of 1 183 AP patients were enrolled finally, of which 129 (10.9%) developed SAP. Among the patients from the First Affiliated Hospital of Soochow University and Changshu Hospital Affiliated to Soochow University, there were 786 patients in the training set and 197 in the validation set; 200 patients from the Second Affiliated Hospital of Soochow University were used as the test set. Analysis of all three datasets showed that patients who advanced to SAP exhibited pathological manifestation such as abnormal respiratory function, coagulation function, liver and kidney function, and lipid metabolism. Based on the XGBoost algorithm, an SAP prediction model was constructed, and ROC curve analysis showed that the accuracy for prediction of SAP reached 0.830, the area under the ROC curve (AUC) was 0.927, which was significantly improved compared with the traditional scoring systems including MCTSI, Ranson, BISAP and SABP, the accuracy was 0.610, 0.690, 0.763, 0.625, and the AUC was 0.689, 0.631, 0.875, and 0.770, respectively. The feature importance analysis based on the XGBoost model showed that the top ten items ranked by the importance of model features were admission pleural effusion (0.119), albumin (Alb, 0.049), triglycerides (TG, 0.036), Ca2+ (0.034), prothrombin time (PT, 0.031), systemic inflammatory response syndrome (SIRS, 0.031), C-reactive protein (CRP, 0.031), platelet count (PLT, 0.030), lactate dehydrogenase (LDH, 0.029), and alkaline phosphatase (ALP, 0.028). The above indicators were of great significance for the XGBoost model to predict SAP. The SHAP contribution analysis based on the XGBoost model showed that the risk of SAP increased significantly when patients had pleural effusion and decreased Alb. A SAP prediction scoring system was established based on the machine automatic learning XGBoost algorithm, which can predict the SAP risk of patients within 48 hours of admission with good accuracy.

Dilemma In Superiority: Rank, Grandiosity, And Aggression Among Men In Uniform

Aggressive, fierce, and grandiose are the words that are often associated with our men in uniform. Law enforcement officers have enough authority and power that can be used against citizens. They are being severely stigmatized by these controversies, and people are generalizing police organizations. Despite the effort to protect the country, it was reciprocated with prejudice and misconception. The purpose of this study was to determine the level of aggression and grandiosity in terms of sex and rank; also to find out if there is a significant relationship between rank, grandiosity, and aggression. The researchers used purposive sampling to choose the respondents. The participants are policemen who have a position and/or rank in a certain PNP office and are available during the conduct of the on-site survey. A set of research questionnaires were distributed to the Philippine National Police of General Santos City-Camp Lira with a total of 55 participants. The responses were analyzed through the use of SPSS. Multiple regression analysis proves that the variables rank, grandiosity, and aggression are significant, < 0.001, and reject the null hypothesis. Additionally, through deeper analysis, only the variables aggression and grandiosity have a significant relationship.

Sociogenesis in unbounded space: modelling self-organised cohesive collective motion

Maintaining cohesion between randomly moving agents in unbounded space is an essential functionality for many real-world applications requiring distributed multi-agent systems. We develop a bio-inspired collective movement model in 1D unbounded space to ensure such functionality. Using an internal agent belief to estimate the mesoscopic state of the system, agent motion is coupled to a dynamically self-generated social ranking variable. This coupling between social information and individual movement is exploited to induce spatial self-sorting and produces an adaptive, group-relative coordinate system that stabilises random motion in unbounded space. We investigate the state-space of the model in terms of its key control parameters and find two separate regimes for the system to attain dynamical cohesive states, including a Partial Sensing regime in which the system self-selects nearest-neighbour distances so as to ensure a near-constant mean number of sensed neighbours. Overall, our approach constitutes a novel theoretical development in models of collective movement, as it considers agents who make decisions based on internal representations of their social environment that explicitly take into account spatial variation in a dynamic internal variable.