Sort by
Performance of the Classical Model in Feature Selection Across Varying Database Sizes of Healthcare Data

Machine learning is increasingly being applied to medical research, particularly in selecting predictive modelling variables. By identifying relevant variables, researchers can improve model accuracy and reliability, leading to better clinical decisions and reduced overfitting. Efficient utilization of resources and the validity of medical research findings depend on selecting the right variables. However, few studies compare the performance of classical and modern methods for selecting characteristics in health datasets, highlighting the need for a critical evaluation to choose the most suitable approach. We analysed the performance of six different variable selection methods, which includes stepwise, forward and backward selection using p-value and AIC, LASSO, and Elastic Net. Health-related surveillance data on behaviors, health status, and medical service usage were used across ten databases, with sizes ranging from 10% to 100%, maintaining consistent outcome proportions. Varying database sizes were utilized to assess their impact on prediction models, as they can significantly influence accuracy, overfitting, generalizability, statistical power, parameter estimation reliability, computational complexity, and variable selection. The stepwise and backward AIC model showed the highest accuracy with an Area under the ROC Curve (AUC) of 0.889. Despite its sparsity, the Lasso and Elastic Net model also performed well. The study also found that binary variables were considered more crucial by the Lasso and Elastic Net model. Importantly, the significance of variables remained consistent across different database sizes. The study shows that no major variations in results between the fitness metric of the model and the number of variables in stepwise and backward p-value models, irrespective of the database's size. LASSO and Elastic Net models surpassed other models throughout various database sizes, and with fewer variables.

Open Access Just Published
Relevant
Estimating Optimum Length of Stay in a Hospital to Control the Infection Spread during an Epidemic Using Left-Right Truncated Poisson Model

Background: The unprecedented havoc of COVID-19 pandemic stressed medical infrastructure of every affected country. The developing countries were more affected as their already inadequate medical resources were strained further. Material and Methods: In order to estimate the time of onset of recovery through the period of hospitalisation stay, the retrospective data on the number of days that 83 COVID-19 patients stayed in a hospital in New Delhi, India was obtained. A Left-Right Truncated Poisson Distribution Model (LRTPD) was developed to estimate the average number of days that patients had to spend in the hospital before the onset of recovery and they were no longer infected. Left truncation is on the ‘u’ left most classes of the random variable and right truncation is after ‘v’ classes. The parametric estimates of the LOS were validated using the Monte-Carlo method. Results and Conclusion: The models suggested that if appropriate truncation limits (both the data-specific and as per expert advice) are used in case of critical medical emergencies, approximately 90 percent of the patients will be able to get hospital admission, without over-burdening the hospital infrastructure. The median recovery onset time/ Length of stay (LOS) obtained using the Kaplan-Meier estimator was consistent with the results of the parametric modeling and simulation. However, the Kaplan-Meier method overestimated the mean LOS as compared to the parametric methods.

Open Access
Relevant
The Impact of Practical Skills on Improving the Servicemen’s Preparedness to Act in Case of Radiation Contamination of the Area

The servicemen’s practical skills to respond to threats of chemical, biological, radiological and nuclear attacks, as well as the ability to make effective decisions are necessary for the implementation of effective targeted actions in the face of military threats. The aim of the article is to identify the impact of servicemen’s decision-making skills on their preparedness to act in case of radiation contamination of the area as well as an analysis of the opportunities of skills development in the educational simulation environment. The research employed such empirical methods as: educational experiment, testing, survey, quantitative assessment, and qualitative analysis. The study of causal relationships between servicemen’s decision-making skills under Contaminated Remains Mitigation System CRMS conditions and their preparedness to act under conditions of radiation contamination made it possible to identify a set of decision-making skills that affect high, medium and low servicemen’s preparedness to act under the chemical, biological, radiological, and nuclear (CBRN) attacks. The authors developed and tested a virtual reality training simulator for training decision-making skills in a simulated environment of potential threats using the Zaporizhzhia Nuclear Power Plant (NPP) situation as an example. The results of the assessment of students’ knowledge after the educational experiment showed that simulation training in virtual reality was more effective than training using educational video content. The students of the experimental group (EG) showed a 13.2 points better result (90.6 points) in decision-making accuracy than the students of the control group (CG) (77.4 points).

Open Access
Relevant
Influence of Stress Factors on the Development of Post-Traumatic Stress Disorder in Children: Risk Factors

In the current conditions of the ongoing war in Ukraine, which began in February 2022, Ukrainian children might develop post-traumatic stress disorder (PTSD) due to stress factors caused by the conflict between Ukraine and Russian-backed separatist forces, along with the Russian military. In particular, the relationship between reduced emotional intelligence and the development of post-traumatic stress disorder in children has become more relevant. This study aimed to assess structural and cognitive changes in children with PTSD and their relationship to depression, anxiety, and event segmentation. The study methods included clinical interviews (CAPS-CA-5 scale), neuropsychological tests (short-term and long-term memory tests), self-assessment questionnaires (standardised CDI, RCADS and SCAS-Child scales), and a single-shotMRI. The results showed that patients with post-traumatic stress disorder had reduced hippocampal volume (p=0.018) and the volume of cingulate cortex isthmus (p=0.026). Diffusion in the cerebellum-hippocampal tract was reduced (p=0.014). The level of depression was positively correlated with hippocampal volume (r=0.32, p=0.021) and anxiety with the volume of cingulate cortex isthmus (r=0.26, p=0.048). These results emphasise the importance of the relationship between structural changes and levels of depression and anxiety in patients with PTSD. Prospects for further research are based on the study of the long-term effects of psychotherapeutic interventions aimed at improving cognitive function in patients with posttraumatic stress disorder.

Open Access
Relevant
The Chronic Progressive Repeated Measures (CPRM) Model for Clinical Trials Comparing Change Over Time in Quantitative Trait Outcomes

Repeated measures analysis is a common analysis plan for clinical trials comparing change over time in quantitative trait outcomes in treatment versus control. Mixed model for repeated measures (MMRM) assuming an unstructured covariance of repeated measures is the default statistical analysis plan, with alternative covariance structures specified in the event that the MMRM model with unstructured covariance does not converge. We here describe a parsimonious covariance structure for repeated measures analysis that is specifically appropriate for longitudinal repeated measures of chronic progressive conditions. This model has the parsimonious features of the mixed effects model with random slopes and intercepts, but without restricting the repeated measure means to be linear with time. We demonstrate with data from completed trials that this pattern of longitudinal trajectories spreading apart over time is typical of Alzheimer’s disease. We further demonstrate that alternative covariance structures typically specified in statistical analysis plans using MMRM perform poorly for chronic progressive conditions, with the compound symmetry model being anticonservative, and the autoregressive model being poorly powered. Finally, we derive power calculation formulas for the chronic progressive repeated measures model that have the advantage of being independent of the design of the pilot studies informing the power calculations. When data follow the pattern of a chronic progressive condition. These power formulas are also appropriate for sizing clinical trials using MMRM analysis with unstructured covariance of repeated measures.

Open Access
Relevant
The Relationship between Traumatic Experiences, the Prevalence of Social Anxiety and Insecure Attachment among University Students

University students face unique challenges and are considered a vulnerable population, making it crucial to understand the impact of trauma on their mental health. This study aimed to investigate the associations between traumatic experiences, the prevalence of social anxiety, and insecure attachment among MSU students. The present study adopted a quantitative research approach using the Trauma Screening Questionnaire (TSQ), the DSM-5 Severity Rating of Social Anxiety Disorder (SAD-D), moreover, for the purpose of assessing PTSD, the Vulnerable Attachment Styles Questionnaire (VASQ), Social Anxiety Disorder Severity, and Insecure Attachment, respectively. A total of 406 respondents participated in the research. Through descriptive analysis, data were collected using three different assessments, revealing that 67% of the students were identified as having a high risk of post-traumatic stress disorder (PTSD), while 6.9% experienced severe social anxiety, which was relatively low compared to the total number. Additionally, 87% of the students displayed a high level of insecure attachment. In order to test the research hypotheses, Pearson correlation analysis, linear regression analysis and path analysis were conducted in this study. The study's findings demonstrated that there was a significant correlation between traumatic experiences and insecure attachment and a non-significant correlation between traumatic experiences and social anxiety. Additionally, traumatic experiences had a significant positive effect on insecure attachment but did not significantly affect social anxiety. Lastly, traumatic experiences did not significantly affect insecure attachment through social anxiety or traumatic experiences through social anxiety.

Open Access
Relevant