International Journal of Statistics in Medical Research

Multiple Mean Comparison for Clusters of Gene Expression Data through the t-SNE Plot and PCA Dimension Reduction

Yiwen Cao — 2025-01-22

This paper introduces a novel methodology for multiple mean comparison of clusters identified in gene expression data through the t-distributed Stochastic Neighbor Embedding (t-SNE) plot, which is a powerful dimensionality re- duction technique for visualizing high-dimensional gene expression data. Our approach integrates the t-SNE visualization with rigorous statistical testing to validate the differences between identified clusters, bridging the gap between exploratory and confirmatory data analysis. We applied our methodology to two real-world gene expression datasets for which the t-SNE plots provided clear separation of clusters corresponding to different expression levels. Our findings underscore the value of combining the t-SNE visualization with multiple mean comparison in gene expression analysis. This integrated approach enhances the interpretability of complex data and provides a robust statistical framework for validating observed patterns. While the classical MANOVA method can be applied to the same multiple mean comparison, it requires a larger total sample size than the data dimension and mostly relies on an asymptotic null distribution. The proposed approach in this paper has broad applicability in the case of high dimension with small sample sizes and an exact null distribution of the test statistic.

Objective: Propose a two-step approach to analysis of gene expression data.

Gene expression data usually possess a complicated nonlinear structure that cannot be visualized under simple linear dimension reduction like the principal component analysis (PCA) method. We propose to employ the existing t-SNE approach to dimension reduction first so that clusters among data can be clearly visualized and then multiple mean comparison methods can be further employed to carry out statistical inference. We propose the PCA-type projected exact F-test for multiple mean comparison among the clusters. It is superior to the classical MANOVA method in the case of high dimension and relatively large number of clusters.

Results: Based on a simple Monte Carlo study on a comparison between the projected F-test and the classical MANOVA Wilks’ Lambda-test and an illustration of two real datasets, we show that the projected F-test has better empirical power performance than the classical Wilks’ Lambda-test. After applying the t-SNE plot to real gene expression data, one can visualize the clear cluster structure. The projected F-test further enhances the interpretability of the t-SNE plot, validating the significant differences among the visualized clusters.

Conclusion: Our findings suggest that the combination of the t-SNE visualization and multiple mean comparison through the PCA-projected exact F-test is a valuable tool for gene expression analysis. It not only enhances the interpretability of high-dimensional data but also provides a rigorous statistical framework for validating the observed patterns.

The Effect of Interpersonal Communication on Prevention Behavior of Early Hypertension among Student at SMAN 6 and SMAN 19 Bone

Andi Alifah Aulya Sultan — 2025-01-27

Background: Hypertension is a health issue that is not only experienced by adults but can also develop during adolescence. This condition often continues into adulthood, with essential hypertension in adults frequently stemming from habits and risk factors that emerge during adolescence. Centers for Disease Control and Prevention (CDC) 2023 revealed that one in every 25 adolescents aged between 12 to 19 years old is diagnosed with hypertension. Among adolescents diagnosed with hypertension, 10% were found to have a prior history of prehypertension.

Objective: This study aims to determine the effect of interpersonal communication on early hypertension prevention behavior among students of SMAN 6 and SMAN 19 Bone.

Materials and Methods: The research design used was Quasi Experiment with pretest-posttest control group design. 110 grade 11 students made up the study population. They were split into two groups: the experimental group, which got an interpersonal communication intervention (n=55), and the control group, which received counseling (n=55). This study was carried out at SMAN 6 and SMAN 19 Bone. Simple random sampling was the method of sampling employed in this study, and a questionnaire was utilized as the research tool to gauge students' knowledge, attitudes, and action both before and after they received the intervention, which had been validated and proven to be reliable. Wilcoxon and Mann-Whitney tests were used for both univariate and bivariate data analysis.

Results: This study showed significant differences in knowledge, attitudes, and actions in the experimental group regarding hypertension prevention behaviors, with p-values for knowledge (p=0.017), attitude (p=0.000), and action (p=0.002).

Conclusion: The interpersonal communication approach applied in the intervention proved to have an influence on hypertension prevention behavior, including knowledge, attitudes, and actions in students of SMAN 6 and SMAN 19 Bone.

Assessing the Impact of Human and Technological Factors on Hospital Management Information System Utilization: A Case Study at Hospital X In Padang City Indonesia

Tosi Rahmaddian — 2025-02-05

This study examines the application of the HOT-Fit method, which evaluates the relationship between Human, Technology, and Net Benefit components within the Hospital Management Information System (HMIS) at Hospital X Padang. The Human component is assessed based on system usage and user satisfaction, while the Technology component is analyzed through information quality, service quality, and system quality. This study employs a quantitative crosssectional design, with the research population comprising all active users of the HMIS application at Hospital X Padang, including employees from various departments interacting with the system. The research aims to determine the extent to which the Human and Technology components influence the Net Benefit of the HMIS and to explore these relationships in greater depth. The findings reveal a significant relationship between the Human component and the Net Benefit, as well as between the Technology component and the Net Benefit of the HMIS. Among the factors examined, Technology emerges as the most dominant factor affecting the Net Benefit of the system. These results provide valuable insights for optimizing the implementation and impact of HMIS in healthcare settings.

Comparative Analysis of Kolmogorov-Inspired CNN and Traditional CNN Models for Pneumonia Detection: A Study on Chest CT Images

Muhammet Sinan Basarslan — 2025-02-05

Aim: In this study, our goal is to compare the effectiveness of Kolmogorov Inspired Convolutional Neural Networks (KAN) with traditional Convolutional Neural Networks (CNN) models in pneumonia detection and to contribute to the development of more efficient and accurate diagnostic tools in the field of medical imaging.

Methods: Both models are structured with the same layers and hyperparameters to ensure a fair comparison of their performance. For a robust evaluation, the relevant dataset was divided into 80% for training and 20% for testing.

Results and Conclusion: Performance metrics of KAN; 95.2% sensitivity, 97.6% specificity, 94.1% precision, 96.9% accuracy (Acc), 0.9466 F1 score (F1) and 0. 9251 Matthews Correlation Coefficient (MCC), while the CNN model was found 92.5%, 96.4%, 91.2%, 95.3%, 0.9188 and 0.8858 for the same criteria, indicating that KAN outperformed. This comparison emphasizes that KAN has the potential to be a more effective model for pneumonia detection in chest CT images.

Predictors of Type-2 Diabetes Self-Screening: The Impact of Health Beliefs Model, Knowledge, and Demographics

Ahmed S. Alamer — 2025-02-13

Background: Diabetes mellitus (DM) is a global health concern, and the intention to undergo diabetes self-screening among patients varies based on demographics and the Health Belief Model (HBM).

Objective: This study aimed to identify the factors associated with the intention to engage in DM self-screening.

Methods: This study included 404 participants with a 99% response rate. Saudi Arabian residents from the Jazan region, all diagnosed with type 2 diabetes, were enrolled. A validated, Arabic-translated, and structured questionnaire was used to collect data on demographics, family history, chronic disease status, DM knowledge, HBM constructs, and DM screening behavior. The study methods adhered to the STROBE Checklist for clear and reliable reporting.

Results: The study found that 24.5% of the participants were in the 35-44 age group and 67.3% were male. Regarding education, 52.2% had university-level education and 79.7% had no family history of DM. Among the participants, 62.1% reported no chronic disease. The mean knowledge score was 6.44 (SD = 2.01). The study revealed that 56.9% of the respondents intended to engage in DM screening. Factors associated with intention included age (65 and over had lower odds), gender (females had slightly higher odds), and education (school qualification had higher odds). Family history and chronic disease status did not significantly affect intention. Among the HBM constructs, higher perceived susceptibility increased the odds, higher perceived severity decreased the odds, and perceived benefits and barriers had no significant associations with intention.

Conclusions: This study provides valuable insights into the factors influencing the intention to engage in DM self-screening among diabetic patients. This understanding can guide targeted interventions to promote DM self-screening and enhance diabetes care outcomes.

Prevalence of Depression among Women Using Hormonal Contraceptive Use: Insights from a Hospital-Based Cross-Sectional Study

Ali Hassan Khormi — 2025-02-17

Background: Hormonal contraceptives (HC) serve as a key component in managing premenopausal symptoms and controlling birth rates. However, mood-related side effects, ranging from minor disturbances to severe clinical depression, are the primary reasons for discontinuation.

Objective: To assess the prevalence of depression among women who use hormonal contraceptive methods. Additionally, the study aims to explore the association between specific types of contraceptives—such as oral pills, implants, injectables—and the prevalence of depression.

Methods: From October 2023 to October 2024, a total of 1500 women between the ages of 21 and 45 who currently take hormonal contraception participated in this hospital-based cross-sectional study, which was carried out at the tertiary care hospital at King Fahd Central Hospital's outpatient gynecology clinic.

Results: The most frequent age categories were from 26 to 40 years (85.7%). The majority of the studied cases were non-lean (82.6%). Most of the cases had parity from 1 to 4 (97.1%). Women were mainly of a low social class (77.1%). Social problems were found in (21.8%). Hypertension and diabetes mellitus were in 4.9% and 3.2% respectively. The most frequent contraceptive method were OCPs (40.3%), followed by POPs (31.2%), then subdermal implants (16.3%), injectable (8.6%), hormonal IUD (2.2%) and patches (1.4%). Most of the studied women used such method from 3 to 6 years (88.2%). Prevalence of depression among the studied cases was (8.7%; CI: 7.3%–10.2%). Obese individuals demonstrated a significantly higher prevalence of depression (11.5%) compared to overweight (8.5%) and lean individuals (5.0%), with a statistically significant association (p=0.015). Additionally, obese participants were more likely to have diabetes mellitus (27.1%), face social issues (21.8%), and belong to a low socioeconomic class (77.1%). Regarding contraceptive types, depression was notably less common among women using combined oral contraceptives (COCs) and progesterone-only pills (POPs), with rates of 4.6% and 4.5%, respectively. In contrast, higher rates of depression were observed in users of subdermal implants (19.2%), injectables (18.6%), hormonal IUDs (18.2%), and hormonal patches (19.0%) (p<0.001). The duration of contraceptive use also played a significant role, with depression rates increasing progressively from 2.8% for women using contraceptives for 1–2 years to 3.7% for 3–4 years and 12.7% for 5–6 years. The highest rate of depression, 37.7%, was observed among women using hormonal contraceptives for seven or more years (p<0.001)

Conclusion: Given the observed associations between certain hormonal contraceptives, prolonged use, and elevated depression rates, clinicians should adopt a proactive approach in assessing patients’ mental well-being, especially for women with additional risk factors like high BMI, socioeconomic challenges, or chronic conditions such as diabetes. Screening tools like the PHQ-9 should be routinely used during consultations to monitor for early signs of depression, allowing for timely intervention if needed.

Understanding Thrombocytopenia in the Obstetric Population: A Study from a Tertiary Care Center

Supriya Jagdale — 2025-02-18

Background: Thrombocytopenia in pregnancy is a common condition with diverse etiologies, ranging from benign causes such as gestational thrombocytopenia (GT) to more serious conditions like preeclampsia and immune thrombocytopenic purpura (ITP). The clinical implications of thrombocytopenia during pregnancy include potential maternal and fetal complications, highlighting the importance of early detection and appropriate management.

Objective: To evaluate the incidence, causes, clinical outcomes, and complications of thrombocytopenia in pregnancy at a tertiary care hospital.

Methods: This retrospective cohort study included 130 pregnant women who were diagnosed with thrombocytopenia during their antenatal care between 2020 and 2021. Data on demographics, etiology, severity of thrombocytopenia, and maternal and fetal outcomes were collected and analyzed.

Results: The incidence of thrombocytopenia in pregnancy was found to be 3.85%. The most common causes were gestational thrombocytopenia (48.48%), preeclampsia (18.18%), and anemia (27.27%). Mild thrombocytopenia (<100,000/µL) was the most frequent severity (68.18%), with severe thrombocytopenia (<50,000/µL) observed in 6.06% of cases. Maternal complications included postpartum hemorrhage (10.60%) and incision site oozing (7.57%). Fetal outcomes included intrauterine growth restriction (12.12%) and birth asphyxia (7.57%). Most cases were diagnosed in the second trimester, and a significant proportion (56.06%) were in primigravida women.

Conclusion: Thrombocytopenia in pregnancy is predominantly mild, with gestational thrombocytopenia being the most common cause. Although the condition generally carries a good prognosis, associated complications such as postpartum hemorrhage and adverse fetal outcomes underscore the need for careful monitoring. Early diagnosis and individualized management are essential to minimize risks for both mother and child.

A Conceptual Model of Sustainable Technology Use: The Role of Confirmation and Perceived Usefulness in the Hospital X Management Information System in Padang

Nurmaines Adhyka — 2025-03-02

Background and Objective: The adoption and use of Management Information Systems (MIS) in healthcare settings, like Hospital X in Padang, are crucial for improving operational efficiencies and patient care. Task-Technology Fit (TTF) measures how well technology supports its intended tasks and significantly influences user satisfaction and system use continuity. Key factors include Confirmation, assessing post-adoption user expectations, and Perceived Usefulness (PU), evaluating job performance enhancement. This study explores TTF's impact on Continuance Intention (CI), mediated by Confirmation and PU, within Hospital X's MIS context.

Methods: Data were gathered from staff at H.B. Saanin Mental Hospital, one of West Sumatera's public hospitals. A total of 158 questionnaires were distributed, with 150 deemed analyzable using structural equation modeling.

Result: The study finds no statistically significant relationship between TTF and PU. However, a marginally significant relationship between TTF and Confirmation suggests modest evidence that alignment between tasks and technology influences users' confirmation of their expectations. Notably, PU does not directly impact CI within Hospital X's MIS, nor does Confirmation significantly affect users' intention to continue using the system. Overall, the direct influence of technology-task alignment on users' intention to continue using MIS is inconclusive in this study context.

Conclution: This study reveals complex relationships among TTF, Confirmation, PU, and CI within Hospital X's MIS framework. Despite the theoretical significance of TTF and Confirmation, their direct impacts on PU and users' intention to continue system use are not statistically significant. These findings emphasize the ongoing need to evaluate and adapt MIS strategies to better align with user needs and ensure sustained effectiveness in healthcare operations.

Optimizing Sample Size for Accelerated Failure Time Model in Progressive Type-II Censoring through Rank Set Sampling

Ibrahim Alliu — 2025-03-02

Survival data is a type of data that measures the time from a defined starting point until the occurrence of a particular event, such as time to death from small cell lung cancer after diagnosis, Length of time in remission for leukemia patients, Length of stay (i.e., time until discharge) in hospital after surgery. The accelerated failure time (AFT) models are popular linear models for analyzing survival data. It provides a linear relationship between the log of the failure time and covariates that affect the expected failure time by contracting or expanding the time scale. This paper examines the performance of the Rank Set Sampling (RSS) on the AFT models for Progressive Type-II censoring-survival data. The Ranked Set Sampling (RSS) is a sampling scheme that selects a sample based on a baseline auxiliary variable for assessing survival time. Simulation studies show that this approach provides a more robust testing procedure, and a more efficient hazard ratio estimate than simple random sampling (SRS). The lung cancer survival data are used to demonstrate the method.

Risk Factors of Physical Condition of House and Clean and Healthy Living Behavior (PHBS) to Tuberculosis in Kaluku Bodoa Health Center Area, Makassar City

Zaenab — 2025-03-02

Background: Tuberculosis remains the 10th leading cause of death globally, accounting for approximately 1.3 million fatalities. The physical conditions of a house, including ventilation, humidity, temperature, occupancy density, lighting, and Clean and Healthy Living Behavior (PHBS), are crucial factors that should be considered in relation to TB incidence.

Objective: This study aims to analyze the relationship between house physical conditions and PHBS with the incidence of TB in the working area of the Kaluku Bodoa Public Health Center, Makassar City.

Methods: This study employed an observational analytic design with a cross-sectional approach. The sample size for the study comprised 150 respondents. Data were processed using univariate analysis, presented in tables, and further analyzed descriptively and bivariately using the Chi-square test to determine the relationship between house physical conditions and CHLB with TB incidence in the working area of the Kaluku Bodoa Public Health Center, Makassar City.

Results: There was a significant relationship between ventilation, lighting, occupancy density, and PHBS with TB incidence in the Kaluku Bodoa Public Health Center, Makassar City. At the same time, temperature and humidity were found to have an insignificant effect on TB incidence.

Conclusion: The findings of this study can be used to guide government policies aimed at improving the quality of life for individuals with TB. Environmental health officers can implement intensive programs emphasizing the importance of handwashing, maintaining cleanliness, and ensuring proper ventilation to reduce the risk of TB transmission.

Chronic kidney Disease Classification through Hybrid Feature Selection and Ensemble Deep Learning

N. Yogesh — 2025-03-03

Diagnosing and treating at-risk patients for chronic kidney disease (CKD) relies heavily on accurately classifying the disease. The use of deep learning models in healthcare research is receiving much interest due to recent developments in the field. CKD has many features; however, only some features contribute weightage for the classification task. Therefore, it is required to eliminate the irrelevant feature before applying the classification task. This paper proposed a hybrid feature selection method by combining the two feature selection techniques: the Boruta and the Recursive Feature Elimination (RFE) method. The features are ranked according to their importance for CKD classification using the Boruta algorithm and refined feature set using the RFE, which recursively eliminates the least important features. The hybrid feature selection method removes the feature with a low recursive score. Later, selected features are given input to the proposed ensemble deep learning method for classification. The experimental ensemble deep learning model with feature selection is compared to Support Vector Machine (SVM), Logistic Regression (LR), and Random Forest (RF) models with and without feature selection. When feature selection is used, the ensemble model improves accuracy by 2%. Experimental results found that these features, age, pus cell clumps, bacteria, and coronary artery disease, do not contribute much to accurate classification tasks. Accuracy, precision, and recall are used to evaluate the ensemble deep learning model.

The Effectiveness of the SOBUMIL mHealth App in Enhancing Early Detection of Pregnancy Complications in Bogor Regency, Indonesia

Bintang Petralina — 2025-03-10

Background: Global and national efforts are underway to reduce maternal mortality. Empowering pregnant women enables health decision-making and early detection of pregnancy complications. Developing applications related to pregnancy potentially improves women's behavior in preventing pregnancy complications.

Objective: This study aimed to explore the influence of SOBUMIL (Sobat Ibu Hamil), an android-based application on pregnant women's empowerment for early detection of complications.

Methods: A quasi-experimental study was conducted in the Bogor Regency, Indonesia. Study participants were pregnant women residing in two primary health care in their second and third trimesters. Pregnant women were excluded if they were disabled or unable to read and write. A total sample of 350 was calculated using the Lemeshow sample formula, which included an intervention and control group.

Results: Overall, we found a statistically significant positive effect of SOBUMIL application in all pregnant women's empowerment parameters to detect pregnancy complications early in Bogor Regency (p<0.001).

Conclusion: This study confirms the positive influence of the SOBUMIL application in empowering pregnant women for early detection of pregnancy complications. This underscores the potential of mobile health interventions to enhance knowledge, attitudes, and abilities, enabling independent monitoring and addressing of pregnancy-related risks, ultimately improving maternal healthcare outcomes.

The Effect of Emotional Regulation for the Successful Treatment of Emotional Dependence in Young People

Serhii Lobanov — 2025-03-25

Regulation of one’s own emotional state is of great importance for a person’s mental health. The issue under research is related to determining emotional regulation approaches for the success of the treatment of emotional dependence in young people. Methods. It was possible to achieve the set goal based on the use of methods of analysis, observation, the Spann-Fischer Codependency Scale, and the Student’s coefficient. The emotional regulation approaches developed by the authors included social recovery, analysis of someone else’s problem and behaviour, problem solving, and art therapy. Results.It was found that the therapy had a positive effect on the respondents, enabling them to primarily develop the self-confidence skills (96%). Also, to develop a lack of need for constant approval (92%), and consideration of their own interests (93%). It was found that the level of the respondents’ emotional dependence decreased to a low level (84%) from the beginning of the study. The respondents noted that art therapy (53%) and socialization (47%), which became the basis of the treatment approaches, had almost the same positive effect. Conclusions.The practical significance of the article is related to the possibility of using effective approaches to regulating emotional dependence in young people. The research prospects will be aimed at comparing the impact of the developed approaches to regulating emotional dependence in young people and middle-aged people.

Policy Innovation in Healthcare: Exploring the Adoption and Implementation of Telemedicine

Abanibhusan Jena — 2025-03-25

Background: Telemedicine has emerged as a transformative solution in healthcare, offering improved accessibility and efficiency. However, its widespread adoption remains influenced by policy frameworks, digital infrastructure, and financial sustainability. This study examines the role of policy innovation in telemedicine adoption and implementation, assessing regulatory impact, technological readiness, and reimbursement structures.

Methods: A cross-sectional survey design with a mixed-methods approach was employed, integrating quantitative surveys and qualitative interviews. Data were collected from healthcare policymakers, administrators, physicians, and technology developers across hospitals, clinics, and telemedicine service providers. Logistic regression and chi-square tests were conducted to analyze key predictors of telemedicine adoption, including regulatory support, digital infrastructure, and reimbursement policies. A total of 400 participants were surveyed, and 25 stakeholders were interviewed to analyze key predictors of telemedicine adoption.

Results: The findings indicate that institutions with clear licensing regulations and policy support exhibited significantly higher telemedicine adoption rates (OR = 2.15, p = 0.004). Standardized reimbursement policies positively influenced adoption rates (χ² = 14.91, p = 0.008). Digital infrastructure readiness, including broadband connectivity and EHR interoperability, was strongly associated with increased telemedicine utilization (OR = 2.31, p = 0.005). Major barriers included regulatory fragmentation, financial constraints, and technological literacy gaps.

Conclusion: Policy innovation, digital infrastructure investments, and structured reimbursement models are critical for telemedicine expansion. Addressing regulatory inconsistencies and financial limitations will enhance adoption. Future research should explore long-term policy impacts and AI integration in telemedicine.

Boruta Feature Selection and Deep Learning for Alzheimer’s Disease Classification

S. Ramu — 2025-03-25

Alzheimer’s Disease (AD) is a progressive neurodegenerative disorder characterized by cognitive decline, memory impairment, and functional deterioration. The early and accurate classification of AD is crucial for timely intervention and management. This study utilizes the Boruta feature selection method to identify the most relevant features for AD classification, selecting the top 15 features based on importance ranking. Three machine learning models—Deep Neural Networks (DNN), Long Short-Term Memory Networks (LSTM), and Support Vector Machines (SVM)—were evaluated using accuracy, precision, recall, and F1-score as performance metrics. The LSTM model demonstrated the highest accuracy (89.30%), outperforming DNN (88.14%) and SVM (84.19%), owing to its capability of capturing temporal dependencies in inpatient data. Results indicate that deep learning models offer superior performance compared to traditional machine learning approaches in AD classification. The study emphasizes the importance of cognitive, lifestyle, and metabolic features in AD diagnosis while acknowledging limitations such as dataset constraints and model interpretability. Future research should improve explainability, incorporate multi-modal data, and leverage real-time monitoring techniques for enhanced AD detection.

A Choice of Performance Metrics for Evaluating Predictive Accuracy of Survival Models

Kumur John Haganawiga — 2025-03-25

This research critically assessed the predictive accuracy of parametric survival models (Weibull, Exponential, Log-logistic, and Gompertz) against penalized Cox PH models (Ridge, Lasso, and Elastic Net) using both simulated data (sample sizes of 100, 200, and 1000) and real-world data from the Nigerian Demographic and Health Survey (NDHS). The findings showed that parametric models, particularly the Weibull and Log-logistic models, consistently outperformed the others, achieving the highest Concordance Index (C-index) and the lowest Mean Absolute Error (MAE) and Mean Squared Error (MSE), indicating superior discrimination and calibration. In contrast, penalized Cox models underperformed, especially with a larger number of covariates, and the Gompertz model exhibited poor predictive performance under all conditions. Notably, parametric models remained stable and consistent even with smaller sample sizes and high-dimensional, complex data. These results highlighted the reliability of parametric models in survival analysis, particularly in small-sample and high-dimensional settings, offering key insights to inform future infant and child health research.