INTRODUCTION: Alzheimer's disease (AD) is a progressive neurodegenerative disorder and the leading cause of dementia. Early diagnosis is vital. We developed an interpretable machine learning (ML) model for early AD prediction using open clinical data.METHODS: Data from 2149 adults (60–90 years) were obtained from Kaggle. After preprocessing and feature engineering, tree‐based models were trained. A stacking ensemble model combining Gradient Boosting and XGBoost was trained, with Logistic Regression as the meta‐learner. SHapley Additive exPlanations (SHAP) provided interpretability. Performance was measured by accuracy, precision, recall, F1 score, ROC and AUC.RESULTS: The stacked ensemble achieved 97% accuracy (AUC 0.97), with 0.97 precision, 0.94 recall, and 0.96 F1 score for AD. SHAP identified memory complaints, Mini‐Mental State Examination (MMSE), functional assessment, behavioral symptoms, cholesterol, and lifestyle factors (activity, diet, sleep) as top predictors.CONCLUSION: The ensemble model, enhanced by SHAP analysis, provides accurate and interpretable AD risk predictions with potential applicability in future clinical decision support systems.Highlights: Developed an ensemble machine learning (ML) model for early Alzheimer's disease (AD) prediction. Achieved 97% accuracy using stacked XGBoost and Gradient Boosting. SHapley Additive exPlanations (SHAP) analysis identified key cognitive and lifestyle‐related risk factors. Model interprets AD risk using explainable artificial intelligence (AI) for clinical applicability. Utilized open‐access dataset to ensure reproducibility and transparency.

The integration of digital technologies into healthcare systems can significantly improve human resource management, especially in low- and middle-income countries (LMICs) like Bangladesh. Despite policy-level efforts such as the rollout of the Human Resource Information System (HRIS), adoption remains uneven and under-studied. This study aimed to identify barriers to HRIS adoption and propose strategies to strengthen digital transformation in human resource management within Bangladesh’s healthcare sector. A cross-sectional mixed-methods study was conducted from March to June 2023 across five districts- Lakshmipur, Feni, Noakhali, Comilla, and Cox’s Bazar- selected for their varying levels of digital readiness. Quantitative data were gathered from 320 h and IT professionals using a structured questionnaire. Qualitative insights were obtained through 10 in-depth key informant interviews with hospital directors, researchers, and policymakers. Quantitative analysis was performed using SPSS version 26, and qualitative data were analyzed thematically using Braun and Clarke’s method. Despite widespread awareness and positive perceptions of HRIS, several critical barriers impeded adoption. These included inadequate training in digital systems, insufficient digital literacy among HR personnel, unreliable internet connectivity, lack of uninterrupted power supply, and limited funding. Furthermore, centralized decision-making, weak inter-ministerial coordination, and lack of motivation among senior leadership emerged as institutional obstacles. Participants emphasized the need for hands-on training, need-based resource allocation, infrastructure investment, and stronger leadership engagement. The convergence of quantitative and qualitative findings underscored the systemic and multidimensional nature of the challenges. Despite progress in digital infrastructure, systemic and organizational barriers continue to hinder the full-scale adoption of HRIS in Bangladesh. Addressing these through targeted interventions is critical for optimizing human resource efficiency and strengthening the health system. Not applicable.

Mental health challenges, particularly depression, anxiety, and stress, are prevalent among university students worldwide. In Bangladesh, public university students face unique stressors, including financial pressure, academic demands, social isolation, and limited access to mental health services. This study aims to identify key risk factors for depression and develop a predictive model to facilitate early intervention. A cross-sectional study was conducted across 15 public universities in Bangladesh, involving 500 students selected through random sampling. Data were collected via an online survey on socio-demographic, academic, and psychological factors. Depression was assessed using a structured questionnaire developed based on PHQ-9 items and aligned with DSM-5 criteria. Descriptive statistics and chi-square tests were used to explore the distribution and associated factors. Machine learning models (Random Forest, SVM, KNN, and Logistic Regression) were applied. Independent t-tests were performed on precision, recall, accuracy, F1-score, and specificity to determine the best model. The prevalence of depression (85.5%) was higher among students, however, there was no statistical significance (p = 0.095) among male students (87.8%) compared to females (81.9%), though. Lower CGPA (p < 0.001), financial stress (p < 0.001), lack of family support (p < 0.001), exposure to ragging (p < 0.001), and frequent session jams (p < 0.001) were significantly associated with depression. Because of precision, recall, accuracy, F1 score and Specificity; the SVM is best model for the dataset to predict depression. Based on SHAP analysis consultancy service, educational system and session jam were found to be the most influential parameters and addressing them can be a game changer. Depression among public university students in Bangladesh are influenced by academic, financial, and institutional factors. Machine learning models, especially SVM offer promising tools for early depression risk prediction, enabling targeted interventions to improve student well-being. The online version contains supplementary material available at 10.1007/s44192-025-00332-0.

End-of-life (EoL) care is the final phase of life (typically the last year when death is imminent), distinct from but related to palliative care, which provides broader support for serious illness. In many low- and middle-income countries, including Bangladesh, end-of-life care remains underdeveloped. Cultural norms, socio-economic disparities, limited palliative care infrastructure, and lack of healthcare provider training further complicate informed decision-making at EoL. This study aimed to evaluate EoL care awareness, preferences, and decision-making factors among critically ill older adult patients in Bangladesh. This cross-sectional study was conducted from September 2024 to February 2025 across eight administrative divisions of Bangladesh, involving 1,270 patients aged ≥ 50 years with chronic or advanced illnesses, including hospitalized patients aged ≥ 18 years with life expectancy < 1 year from private (n = 368), public (n = 439), and community (n = 463) settings. Divisional hospitals provided hospital-based data, while three randomly selected sub-districts per division ensured proportional community representation. Data was collected using structured questionnaires adapted from validated international tools, capturing socio-demographics, and end-of-life (EoL) awareness, preferences, and experiences. Predictors of end-of-life preferences were examined using multiple logistic regression analysis. Only 6.93% had health insurance, with just 1.7% in community settings. Palliative care awareness was highest in private hospitals (70%), followed by public (31%) and community settings (7.1%) (p < 0.01). Advance care planning awareness and documentation were lowest in community patients (p < 0.01). Older adults (≥ 60 years) were more likely to prefer home care (OR = 2.96, p = 0.004), avoid hospitalization (OR = 17.55, p < 0.001), and choose home death (OR = 10.29, p < 0.001). Greater understanding of palliative care (OR = 7.38, p < 0.001) and hospice comfort (OR = 25.26, p < 0.001) strongly predicted documentation of end-of-life preferences. Family openness varied widely (private: 81%, public: 21%, community: 7.1%). Proxy appointment was significantly associated with prior discussions (AOR = 4.11), while trust in healthcare providers reduced the likelihood (AOR = 0.39). Profound disparities in end-of-life awareness and preferences exist across healthcare settings in Bangladesh, driven by socio-economic, cultural, and institutional factors. Efforts must prioritize patient choice through enhanced communication, culturally sensitive advance care planning, and expanding home-based palliative options to honor care preferences at the end of life.

Background: Type 2 diabetes mellitus (T2DM) is a growing public health issue in Bangladesh, projected to affect 13.7 million individuals by 2045. Physical inactivity and poor dietary habits increase the risk of complications. This study examined the effects of physical activity and lifestyle modifications on T2DM management.Methods: A case–control study was conducted at Manikganj District Diabetic Hospital with 334 T2DM patients (aged 45–60 years). The case group (n = 167) followed structured physical activity and dietary modifications, while the control group (n = 167) did not. Data on socio‐demographics, lifestyle and glycaemic markers—fasting plasma glucose (FPG), postprandial plasma glucose (PPG) and HbA1c—were collected at baseline, 3 months and 6 months. Chi‐square tests and ordinal logistic regression models assessed associations between lifestyle factors and glycaemic outcomes.Results: The control group had significantly higher plasma glucose levels, associated with gender (p < 0.001), treatment type (p = 0.004), medical adherence (p = 0.009), food habits (p = 0.007) and BMI (p = 0.005). The case group showed a significant FPG reduction from 10.035 mmol/L to 6.261 mmol/L (p < 0.001), with similar trends for PPG and HbA1c. Males had 1.8 times higher odds of elevated FPG, while poor adherence increased this risk 2.5‐fold.Conclusions: Lifestyle modifications significantly improve glycaemic control in T2DM patients. Strengthening adherence to medical advice and integrating structured interventions into routine care could enhance diabetes management in Bangladesh.