The purpose of this study is to prioritize key environmental sustainability dimensions and indicators within the pharmaceutical sector to support informed decision-making and policy formulation. The study adopts a quantitative methodology using RIDIT analysis to compare multiple sustainability indicators against a reference distribution and derive priority rankings. Survey data collected from 516 respondents were analyzed across environmental impact, environmental sustainability, impact investing, and adoption of green pharmaceutical practices. The findings reveal that environmental sustainability and environmental impact indicators receive the highest priority, indicating strong stakeholder emphasis on long-term ecological outcomes. Impact investing plays a moderate facilitative role, while green practice adoption ranks lowest, highlighting implementation gaps. The study offers practical implications for policymakers and industry managers by identifying areas requiring targeted interventions and resource allocation. The novelty of this research lies in the application of RIDIT analysis to systematically rank sustainability dimensions in the pharmaceutical context, providing a robust and interpretable prioritization framework.

The rapid growth of healthcare information systems has led to an increase in the volume of clinical data stored in electronic medical record systems. This data contains valuable information such as patient demographics, vital signs, diagnoses, and medical procedures, which can be used to support clinical decision-making. Determining the appropriate clinical procedure remains a challenge due to the complexity of patient conditions and the potential need for multiple procedures simultaneously. This study aims to develop a robust model for predicting clinical procedures using a multi-label classification approach based on electronic medical record data. The proposed methodology integrates data preprocessing, feature engineering, followed by the implementation of several Machine Learning methods such as LightGBM, XGBoost, CatBoost, Deep Learning Models and combined using an Ensemble Learning method. Various Ensemble Learning techniques, including Simple Average, Machine Learning Average, and Optimized Weight-based combinations, are applied to improve predictive performance. Evaluation is performed using the Micro F1 score and Macro F1 score to assess overall performance. Experimental results show that individual models achieve competitive performance, with LightGBM, XGBoost, CatBoost outperforming the deep learning approach on tabular clinical data. The Ensemble Learning approach further improved performance, with the Optimized Weight with Grid Search Optimization ensemble achieving the highest Micro F1 score of 0.7788 and Macro F1 score of 0.7674. These results also demonstrate that combining multiple models effectively reduces bias and variance while improving generalization. The small difference between Micro and Macro F1 scores indicates balanced performance across labels. In conclusion, the proposed Ensemble Learning-based multi-label prediction model demonstrates robust capabilities in handling complex clinical data and improving prediction accuracy. This study highlights the importance of model combination strategies rather than increasing model complexity. These findings have practical implications for supporting clinical decision-making systems in healthcare settings.

Despite the growing adoption of artificial intelligence (AI) in healthcare, improving AI frameworks remains essential for effective medical implementation. A major challenge in developing reliable machine learning models is handling imbalanced medical datasets. Although metaheuristic algorithms can improve machine learning performance, their effectiveness on imbalanced medical data has received limited attention. This study evaluates four nature-inspired optimization techniques, bat algorithm (BA), genetic algorithm (GA), improved grey wolf optimization (IGWO), and improved whale optimization (IWO), for enhancing balanced random forest (BRF) classification on imbalanced datasets involving breast tumors, heart failure survival, mammogram microcalcification detection, and diabetes. Based on F1 scores, IGWO achieved the greatest improvements (0.256–0.060), followed by IWO (0.229–0.045), GA (0.214–0.036), and BA (0.203–0.036). Validation results confirmed IGWO’s strong convergence and optimization efficiency, consistently outperforming the other methods. The best performance was achieved by IGWO-BRF in breast cancer diagnosis (F1 = 1.000), while the lowest was observed for BA-BRF in mammogram microcalcification detection (F1 = 0.570). Optimization time depended on both the search strategy and dataset characteristics, with IWO showing the fastest optimization across all datasets. Overall, nature-inspired optimizers significantly improved classification performance on imbalanced medical datasets.

Antenatal care (ANC) is a key determinant of maternal and neonatal health, yet its utilization remains uneven across India. This study examined the socio-demographic determinants of receiving the recommended four or more ANC visits using data from 232,920 women aged 15–49 years from the National Family Health Survey (NFHS-5). Descriptive statistics and multivariable logistic regression were used to identify factors associated with adequate ANC utilization. Overall, 57.9% of women received four or more ANC visits. Higher maternal education, better household wealth, husband's education, and regular mass media exposure significantly increased the likelihood of adequate ANC utilization, whereas rural residence was associated with lower utilization. Women aged 25–39 years were also more likely to complete the recommended ANC visits than younger women. The findings highlight persistent socio-economic and geographical disparities in maternal healthcare utilization and emphasize the need for targeted interventions focusing on education, rural healthcare access, economic empowerment, and health awareness to improve equitable ANC coverage in India.

Cardiovascular disease remains the leading cause of mortality worldwide, driving substantial research investment and necessitating robust evaluation of institutional research performance. This study employed the Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) to evaluate and rank the research performance of 30 leading universities based on their contributions to the top 100 highly cited cardiovascular medicine articles from 2010–2026, extracted from Scopus. Three evaluation criteria were selected: Unique Authors (minimized as a resource input measure), Total Citations (maximized as an impact indicator), and Number of Papers (maximized as a productivity measure), with equal weights (33.3% each) assigned to all criteria. TOPSIS analysis revealed substantial variation in institutional performance, with Harvard University emerging as the top-performing institution (TOPSIS score: 0.6050), demonstrating dominance across all criteria with 228 unique authors, 9,430 citations, and 83 papers, followed by Cleveland Clinic (0.4882), University of Toronto (0.4537), and Mayo Clinic (0.4158). US institutions dominated the ranking, accounting for 27 of 30 universities with an average TOPSIS score of 0.4112, while Canadian institutions achieved an average of 0.4078. The analysis identified an “efficiency gap” between high-citation, low-paper institutions (e.g., UCSF with 87.4 citations per paper and 1.3 authors per paper) and those demonstrating the inverse pattern. The findings highlight the dominance of US institutions, varying institutional strategies (high-volume vs. high-impact approaches), and the concentration of research activity among leading academic medical centers, with the TOPSIS methodology offering a robust framework for multidimensional research evaluation that can inform institutional benchmarking, strategic planning, and resource allocation in cardiovascular medicine.

Health performance in India varies markedly across states and Union Territories, yet composite rankings that summarise this variation are sensitive to how criteria are weighted and how alternatives are aggregated. Subjective weighting and single-method aggregation limit the transparency and reproducibility of existing rankings. This study constructs an objective, reproducible composite ranking of subnational health-system performance using the sixth round of the National Family Health Survey (NFHS-6, 2023-24). Thirty-five states and Union Territories were evaluated on twenty-one curated headline indicators spanning six analytical domains (marriage and fertility, maternal health, child health, child nutrition, adult nutrition and non-communicable disease risk, and women's empowerment). Benefit, cost, and target-based criteria were handled coherently; the two target criteria (total fertility rate and caesarean section rate) were transformed to distance from their clinical optimum. Objective weights were derived with the Method based on the Removal Effects of Criteria (MEREC). Alternatives were ranked with the Alternative Ranking Order Method Accounting for Two-step Normalization (AROMAN), implemented with type-aware normalization. Robustness was assessed through weight perturbation, aggregation-parameter sweeps, an all-eligible-criteria variant, and an imputation sensitivity analysis. Convergent validity was examined against four independent ranking methods, and external validity against independent socioeconomic measures. The two fertility-related target criteria and child marriage received the largest MEREC weights (0.103, 0.099, and 0.075, respectively). Ladakh, Himachal Pradesh, and Mizoram occupied the top three positions, whereas Bihar, Uttar Pradesh, and Tripura occupied the bottom three. The ranking was highly stable: the mean Spearman correlation between the base ranking and 5000 randomly perturbed weight vectors was 0.99, and rank correlations across the aggregation parameter exceeded 0.93. Agreement with the four comparison methods was strong at the policy-relevant extremes (WS rank-similarity coefficient between 0.91 and 0.96) and moderate to strong across the full ordering (Spearman correlation between 0.49 and 0.83). The composite score correlated positively with the socioeconomic enabling environment (Spearman correlation 0.69, p ≤0.001). The integrated MEREC-AROMAN framework produces a transparent and reproducible ranking of Indian subnational health performance that is stable under perturbation and consistent with independent evidence. The persistent disadvantage of the large northern states, alongside the emerging burden of over-medicalisation and adult obesity in the southern states, indicates that a single national policy prescription is unlikely to be adequate.

The Cardiac Health Assessment System for vehicle drivers utilizes the synergy and combined effect arising from the integration of Machine Learning, Internet of Things that will allow continuous monitoring and analysis of the driver's cardiovascular well-being. The current system Primarily consists of detecting Cardiac Arrhythmia, is founded on traditional Electrocardiogram (ECG) monitoring, which requires electrodes being placed on the skin of the patients to register the electrical activities. With the use of smart wearables and ECG sensors, the system can register the activities in real time heart health data when the person is driving and when not driving. With the use of Internet of Things (IoT), the details are seamlessly communicated to a central system. Sophisticated Machine Learning algorithms evaluate these ECG patterns to provide insights to the early detection of any irregularities and risks. The conventional ECG devices are not coupled with the Internet of Things. sensors, limiting the scope of further necessary data, like age, sex, height, and weight. In the case of irregularities, the system provides immediate notifications, helping towards quick intervention and, if required, remote medical consultations. The friendly interface of this system gives drivers access to their health data, as well as personalized suggestions, to cultivate an active health care strategy This unique innovation did not occur in isolation but only improves road safety as far as indicating possible cardiac problems but also helps in creating a data-driven the level of understanding pertaining to the relationship between cardiovascular well-being and driving factors. This research work describes the design of an Internet of Things -assisted ECG monitoring and heart disease diagnosis system specifically for vehicles. The system can acquire real-time ECG signals from wearable biosensors and transmit them. to a cloud platform by means of low-power Internet of Thing modules. The vehicle-monitoring system offers immediate notification to all drivers and subscribed stakeholders when abnormal patterns emerge, improving safety on the roads and allowing for immediate medical intervention. Thus, this concept promotes continuous physiological surveillance, fast risk analysis, and preventive measures for heart-related incidents during driving.

Wearable and Internet of Medical Things devices increasingly support continuous health monitoring, but cloud-dependent analytics remain constrained by latency, connectivity, privacy, and battery requirements. Tiny machine learning shifts inference toward resource-constrained microcontrollers and edge processors; however, its clinical value depends not only on predictive accuracy but also on energy efficiency, real-time responsiveness, and understandable decision logic. This scoping review mapped the evidence on explainability and energy efficiency in TinyML-based health monitoring. Following the Joanna Briggs Institute approach and PRISMA guidance for scoping reviews, Scopus was searched in the title, abstract, and keyword fields for studies published from 2020 to 2025. The search combined TinyML and embedded or edge artificial intelligence terms with wearable or Internet of Medical Things concepts, health-monitoring applications, and explainability or efficiency terms. The supplied export contained 293 records. After screening, 39 reports underwent eligibility assessment and 36 studies were included. Publication activity accelerated sharply, with 18 studies published in 2025. The evidence covered cardiac monitoring, human activity and fall detection, neurological and affective assessment, respiratory monitoring, signal-quality control, glucose sensing, gait analysis, and smart textiles. Convolutional neural networks and hybrid deep models were common, while deployment platforms ranged from microcontrollers to field-programmable gate arrays, application-specific integrated circuits, and neuromorphic hardware. Quantization, pruning, binary or ternary computation, feature reduction, event-driven processing, and local transmission control were frequently used to reduce resource demand. In contrast, only a small minority of studies explicitly evaluated explainability through model-based feature selection, feature importance, or class activation maps. The field is therefore energy-aware but not yet consistently explanation-aware. Future research should adopt standardized hardware reporting, clinician-centered explanation evaluation, external and longitudinal validation, and multiobjective optimization that jointly considers clinical accuracy, energy, latency, memory, robustness, and interpretability.

Healthcare distribution companies generate large volumes of transaction data that can support product-combination decisions. However, conventional Market Basket Analysis often relies on support-based measures, which prioritize frequently purchased products without considering their economic contribution. This study proposes a revenue-based High-Utility Itemset Mining framework to identify economically valuable product combinations from healthcare distribution transaction data. The proposed approach uses transaction-level quantity, unit selling price, discount, and realized line-level transaction value to calculate product utility. FP-Growth is applied as a baseline support-based method, while High-Utility Itemset Mining is used to discover product combinations that exceed a minimum utility threshold. The empirical results show that support-based and utility-based mining produce substantially different insights. Several products with high support contribute relatively low utility, while some lower-support itemsets generate high utility shares. At the 1.00% minimum utility threshold, the proposed framework identifies 484 high-utility itemsets after evaluating 12,660 candidate itemsets and pruning 9,192 candidates using Transaction-Weighted Utility. The findings support cross-selling, product bundling, and inventory prioritization decisions. This study demonstrates that utility-based mining provides more economically relevant product-combination insights than frequency-based Market Basket Analysis alone.

Breast cancer diagnosis using medical imaging remains a critical challenge due to variability in tumor appearance and limitations of single-model prediction systems. While deep learning approaches have yielded encouraging results, concerns with model generalization, robustness, and clinical reliability remain. This study provides a hybrid ensemble deep learning framework for ultrasound-based breast cancer classification, which combines a proprietary Convolutional Neural Network (CNN) with transfer learning models such as VGG16, DenseNet121, and InceptionV3. A total dataset of 5,400 ultrasound images was used. Parameter optimization and preprocessing techniques such as normalization, resizing, and data augmentation were employed. The proposed system uses voting ensemble approaches to combine predictions from individual models. DenseNet121 achieved the highest standalone accuracy (98.2%), whereas the proposed ensemble model demonstrated superior generalization, with an overall accuracy of 96.0%, precision of 96.1%, recall of 96.0%, and F1-score of 95.9%. Soft voting outperformed hard voting in terms of stability and balanced classification performance, whereas hard voting reduced model bias. The data show that ensemble learning increases diagnostic consistency and reliability. The proposed method offers a scalable and robust solution for computer-aided diagnosis systems, with high potential for real-world clinical integration.