Sickle Cell Disease (SCD) is a hereditary blood disorder characterized by abnormally shaped red blood cells, causing anemia, vascular blockages, and severe health complications. Early and accurate diagnosis is essential, but access to reliable diagnostic tools is often limited in resource-constrained settings. This study proposes an automated image segmentation and classification pipeline to improve SCD detection. Using the ErythrocytesIDB dataset, which contains labeled erythrocyte images classified as circular, elongated, or irregular, Gaussian Filter is applied for denoising, followed by OTSU thresholding for segmentation. Data augmentation techniques, including rotation, shifting, and flipping, enhance model robustness and generalization. A Convolutional Neural Network (CNN) integrated with an attention mechanism is developed for accurate erythrocyte classification, and five optimizers: ADAM, SGD, RMSPROP, ADAMAX, and NADAM are systematically evaluated. Validation via 5-fold cross-validation demonstrates that the proposed preprocessing and augmentation steps significantly improve performance. The CNN with attention optimized using NADAM achieves the highest test accuracy of 98.33%, outperforming baseline and recent state-of-the-art models. The proposed pipeline provides a reliable, efficient, and scalable solution for automated SCD detection, particularly suitable for regions with limited access to advanced medical diagnostics.

Prostate cancer is the second most commonly diagnosed cancer worldwide. Prostate biopsy, essential for definitive diagnosis, has evolved significantly with new technologies and techniques. Transrectal ultrasound-guided biopsy (TRUS-Bx) has been the gold standard but carries substantial infectious risks due to rectal mucosal penetration. Rising antibiotic resistance, emerging safety protocols, and novel imaging-guided methods have driven a shift toward safer alternatives. Following PRISMA guidelines, we systematically searched PubMed and PubMed Central for studies published between 2014 and 2025 on prostate puncture complications. Eligible articles included original studies with ≥2 patients, emphasizing infectious complications, antibiotic prophylaxis, and modern innovations. From 639 records screened, 78 met inclusion criteria. Thematic synthesis was used to classify findings into complication types, prophylaxis approaches, and technological advancements. Infectious complications after TRUS-Bx ranged from 0.5 to 9.4% for sepsis and 0.3 to 4.9% for febrile urinary tract infections, largely driven by multidrug-resistant organisms and increased sampling density. Transperineal biopsy (TP-Bx), bypassing rectal flora, consistently reported infection rates <1%. Targeted prophylaxis based on rectal cultures, combination antibiotic regimens (e.g., fluoroquinolone with fosfomycin or ceftriaxone), and adjunct measures such as rectal cleansing significantly reduced post-biopsy infections. Technological innovations such as MRI-ultrasound fusion, robotic-assisted approaches, and PSMA PET/CT-guided techniques improved cancer detection rates (up to 71.8%) while maintaining low complication rates ( < 5%). Emerging non-antibiotic TP protocols and advanced anesthetic techniques further enhanced safety and patient tolerance. Modern evidence supports a paradigm shift toward TP-Bx combined with targeted or multidrug prophylaxis to mitigate infectious risks. Imaging-guided and robotic-assisted techniques enhance diagnostic accuracy and safety but remain limited in resource-constrained settings. TRUS-Bx retains utility where TP access is unavailable; however, adapting infection prevention strategies is critical. Future large-scale trials and cost-effectiveness analyses are needed to optimize biopsy protocols globally.

Clinically, epilepsy manifests as a chronic condition marked by unprovoked, recurrent seizures, plaguing over 70 million individuals with debilitating seizures and life-threatening complications. Approximately 30% of patients with epilepsy do not respond to conventional antiepileptic drugs, indicating the limited efficacy of these medications in controlling seizures universally. Therefore, seizure prediction has become a key factor in enabling timely intervention for epilepsy patients, which can provide crucial time for clinical treatment and preventive measures. This study aimed to propose a lightweight seizure prediction model integrating a residual network (ResNet) with a kernel-enhanced global temporal attention Block (GTA Block). The ResNet extracts electroencephalogram (EEG) features while maintaining gradient stability, and the GTA mechanism constructs full-sequence temporal association matrices to capture the dynamic evolution of EEG patterns. Then a kernel function is embedded into GTA Block for mapping EEG samples into a high-dimensional space in which the distinction between preictal and interictal states is enhanced. The model significantly outperforms existing methods while maintaining a lightweight architecture suitable for embedded systems. With only 1.94 million parameters and an inference time of 0.00207[Formula: see text]s, this lightweight design facilitates real-time deployment on wearable devices, enhancing feasibility for continuous clinical monitoring in resource-constrained settings.

Aim: Early screening for autism spectrum disorder (ASD) using facial images is promising but often limited by small datasets and the lack of deployable models for resource-constrained settings. To develop and evaluate a lightweight framework that combines a multi-scale vision transformer (MS-ViT) with edge optimization for ASD classification from children’s facial images. Methods: We analyzed 2,940 RGB facial images of children obtained from a publicly available Kaggle dataset. Faces were detected, aligned, and cropped (ROI extraction), then normalized; training used standard augmentations. The backbone was an MS-ViT with multi-scale feature aggregation. We performed an 80/20 stratified split (training/testing) and used five-fold cross-validation within the training set for validation (i.e., ~64% training, ~16% validation, and 20% testing per fold). Edge deployment was enabled through post-training optimization. Performance was assessed using accuracy, sensitivity, specificity, AUC-ROC, and per-image inference time. Results: The best configuration (MS-ViT + Edge + Augmented) achieved an accuracy of 96.85%, sensitivity of 96.09%, specificity of 97.92%, and AUC-ROC of 0.9874. On a Raspberry Pi-class device, the model reached ~181 milliseconds per image, supporting real-time screening. Conclusions: The proposed “MS-ViT + Edge + Augmented” framework offers near-state-of-the-art accuracy with low latency on low-power hardware, making it a practical candidate for early ASD screening in clinics and schools. Limitations include dataset size and demographic diversity; prospective clinical validation on larger, multi-site cohorts is warranted.

BackgroundTherapeutic inertia the failure to intensify treatment despite persistent hyperglycemia is a major barrier to optimal management of type 2 diabetes, particularly in low-resource settings.MethodsA hospital-based cross-sectional study was conducted from June 1, 2024, to August 30, 2024. A total of 299 systematically selected patients were included. Data were collected via structured questionnaires and patient medical records. Bivariable and multivariable binary logistic regression analyses were used to identify factors associated with therapeutic inertia. Variables with a p value < 0.25 in the bivariable analysis were included in the multivariable model, and those with a p value < 0.05 were considered statistically significant.ResultsOverall, 67.2% of patients experienced therapeutic inertia. Multivariable analysis identified four independent predictors: lack of health insurance reduced the likelihood of treatment intensification (AOR = 0.177; 95% CI: 0.054–0.576; p = 0.004); management by general practitioners doubled the odds of inertia compared with specialist care (AOR = 2.002; 95% CI: 1.017–3.939; p = 0.045); higher baseline fasting plasma glucose was associated with increased odds of inertia (AOR = 1.008; 95% CI: 1.003–1.013; p = 0.003); and limited availability of point-of-care HbA1c testing substantially increased the risk of inertia (AOR = 8.423; 95% CI: 1.889–37.561; p = 0.005).ConclusionTherapeutic inertia is highly prevalent, affecting 67.2% of ambulatory patients with type 2 diabetes at NEMMCSH. This study highlights critical barriers at patient, provider, and system levels. Interventions such as expanding insurance coverage, enhancing provider training and decision support, implementing prompts for elevated glycemia, and integrating point-of-care HbA1c testing are urgently needed to reduce therapeutic inertia and improve glycemic control in resource-constrained settings.

This study addresses the challenge of accurately quantifying cardiac left ventricle (LV) function, critical for diagnosing cardiovascular diseases. Existing methods typically depend on segmentation-based models that require large annotated datasets, a resource often scarce in the medical field. Moreover, the low inter-class variability and high noise in ultrasound images further complicate the model training. To overcome these limitations, we propose LV-STANet, a segmentation-free model designed to minimize reliance on ground truth annotations while maintaining accuracy and computational efficiency. LV-STANet estimates LV function directly from 2D echocardiogram videos by integrating spatial and temporal features. A spatial encoder captures anatomical features, while a temporal attention module models the dynamic behavior across frames. These components are combined using a weighted aggregation strategy to predict key LV functional parameters: ejection fraction (EF), global longitudinal strain (GLS), and fractional shortening (FS). We evaluate our model on the publicly available EchoNet-Dynamic dataset. LV-STANet achieves a mean absolute error (MAE) of 5.1% for EF, 3.35% for GLS, and 4.95% for FS, demonstrating competitive performance. These results highlight the model' s ability to provide accurate and reliable cardiac function assessment without the need for segmentation, offering a promising direction for clinical deployment in resource-constrained settings.

Radiographers' perspectives on triage systems: Exploring workflow impacts and enhancement opportunities in resource-constrained radiology departments.

A scoping review of climate resilient health system strategies in low-resource settings.

Over the past 34 years, a single transplant team in São Paulo, Brazil, has performed 1560 pediatric liver transplants (PLT)-including 1352 LDLT, 179 DDT, and 29 domino procedures using donors with maple syrup urine disease-achieving outstanding long-term outcomes. In our most recent cohort of 500 PLT (2018-2024), 1- and 5-year patient survival rates were 96.7% and 94.8%, respectively. From 2015 to 2024, our team performed 35.1% of all PLT in Brazil and 45.5% of those in children under 5 years of age, with 98.1% of the latter using LDLT. This milestone highlights not only clinical achievement but also the persistent structural challenges facing PLT in low- and middle-income countries. We outline some key barriers to sustainable and equitable PLT development in Brazil, including underfunding, geographic disparities, lack of outcome transparency, bureaucratic delays, and gaps in transition to adult care. A set of guiding principles is proposed to support national progress and inform similar efforts elsewhere. Our center's experience demonstrates that excellence in PLT is achievable in resource-constrained settings, but long-term success depends on institutional commitment, strategic investment, and national coordination to ensure equitable access for all children.

A 12-hour versus 24-hour magnesium sulfate intravenous regimen in postpartum women with preeclampsia: a randomized clinical trial.

Colorectal disease diagnosis with deep triple-stream fusion and attention refinement.

With the growing challenges faced by visually impaired and elderly individuals, particularly regarding mobility and safety, the development of advanced assistive technologies has become crucial. This paper presents the design and implementation of an Advanced Blind Stick (ABS) leveraging Internet of Things (IoT) technology. The ABS integrates multiple sensors, including ultrasonic sensors for obstacle detection, infrared sensors for fire and pothole/stair detection, a moisture sensor for puddle detection, a GPS module for real-time location tracking, and a thermistor for temperature monitoring. An Arduino microcontroller processes sensor data, providing real-time feedback via a smartphone application. The system also includes an emergency save our soul (SOS) button, allowing users to alert caregivers or emergency services during critical situations. The developed ABS enhances mobility and safety by incorporating advanced features absent in previous models. It uniquely detects puddle, monitors both ambient and user temperature, and retains essential functionalities such as potholes/stair detection, IoT integration, caregiver reporting, SOS, and fire detection. This comprehensive assistive device ensures real-time assistance, improved usability, and greater independence for users. Tested with visually impaired and elderly participants, the ABS demonstrated significant improvements in mobility, safety, and confidence. Its lightweight, cost-effective design makes it particularly beneficial for resource-constrained settings, offering an invaluable tool to enhance the independence and well-being of visually impaired and elderly individuals.

Agriculture, as a critical sector of the Indian economy, is increasingly benefiting from advancements in mobile and cloud-based technologies. This paper presents Fertiflow, a cross platform agricultural application developed using Flutter and Firebase, designed to modernise the digital infrastructure available to farmers. Fertiflow leverages modern software engineering principles to deliver a modular and scalable solution that connects farmers with essential services, including market pricing, government schemes, agency support, and digital payments. The system architecture comprises key application modules, including Home Screen, Agency Screen, Profile Screen, Login Screen, Payment Screen, and dedicated components for Government MSP integration and Farming Tools access. Built on Firebase’s real-time database and authentication services, the app ensures seamless synchronization, secure user management, and efficient data handling in low-connectivity rural environments. This paper examines the user interface design, component-based architecture, and backend integration strategy of the application, as well as the challenges of delivering reliable performance in resource-constrained settings. Additionally, it discusses the broader implications of Fertiflow in advancing rural digital transformation, highlighting how mobile-first software solutions can bridge the gap between traditional farming practices and emerging digital ecosystems. Fertiflow exemplifies the convergence of mobile computing, cloud services, and agricultural informatics, demonstrating how computer science can directly contribute to socio-economic development in rural sectors.

Background: Timely management in the emergency department (ED) is critical for patient safety and quality of care. Prolonged Length of Stay (LOS) in the ED, often linked to delays in patient processing, can lead to poor outcomes, particularly in high-risk populations. This study aims to explore the time-related factors influencing LOS using the Time Range Guidance Model. Methods: A retrospective cross-sectional study was conducted, analysing 377 patient records from June to August 2025. Data were extracted from electronic medical records (EMR), focusing on time intervals for initial assessment, review/consultation, and transfer phases. Multivariate regression was used to identify the factors influencing LOS. Results: The review/consultation time (β = 0.3, p &lt; 0.001) and transfer waiting time (β = 0.356, p &lt; 0.001) were significant predictors of LOS. The model explained 22.6% of the variability in LOS (Adjusted R² = 0.226). Initial assessment time did not significantly correlate with LOS (r = 0.045, p = 0.321). Conclusion: Delays in the review/consultation and transfer phases have a greater impact on ED LOS than the initial assessment. These findings suggest that optimizing these phases can improve ED efficiency and patient outcomes, particularly in resource-constrained settings such as Indonesia.

This study addresses the problem of authorship attribution for Romanian texts using the ROST corpus, a standard benchmark in the field. We systematically evaluate six machine learning techniques — Support Vector Machine (SVM), Logistic Regression (LR), k-Nearest Neighbors (k-NN), Decision Trees (DT), Random Forests (RF), and Artificial Neural Networks (ANN), employing character n-gram features for classification. Among these, the ANN model achieved the highest performance, including perfect classification in four out of fifteen runs when using 5-gram features. These results demonstrate that lightweight, interpretable character n-gram approaches can deliver state-of-the-art accuracy for Romanian authorship attribution, rivaling more complex methods. Our findings highlight the potential of simple stylometric features in resource-constrained or under-studied language settings.

Background: Identifying surgical patients through administrative and clinical data can inform the quality and demand for surgical care. In South Africa, a database exists that comprises data from the public health sector. However, algorithms are lacking to identify surgical procedures like appendectomy in these systems in our setting. Aim: To develop and validate an appendectomy algorithm for use in a South African database. Setting: Data from public hospitals in South Africa were the reference standard and comprised appendectomy and other general surgery procedure controls. The index test was the appendectomy algorithm developed and validated using the provincial database in the country. Methods: A diagnostic test accuracy study was done. The algorithm was developed using four phases: exploration and selection, development, refinement and validation. Data analyses were performed using STATA version 18. Results: The final algorithm comprised two procedures and nine diagnostic codes and reached a sensitivity of 91.3% and a specificity of 96%. Conclusion: Our study is the first to validate an appendectomy algorithm in a low-and middle-income country setting. While not the first globally, it addresses a critical gap in the literature by demonstrating that robust, high-specificity algorithms can be developed in resource-constrained settings. Future research should focus on applying the algorithm to evaluate median delays in accessing care within the public health system. Contribution: This study demonstrates that surgical procedure algorithms can be developed and validated with sufficient sensitivity and specificity using diagnostic and procedure codes for application in a low- and middle-income country setting.

In Sri Lanka’s emerging economy, entrepreneurial engineers who lead their own enterprises face the dual challenge of technical innovation and workforce management. This study investigates how these entrepreneurial engineers maintain discipline within their organizations, an often-overlooked aspect of engineering entrepreneurship. Based on 22 in-depth interviews with founders and leaders of engineering-based businesses across Sri Lanka, the research employs an inductive thematic analysis to uncover key strategies, obstacles, and cultural dynamics that shape disciplinary practices in privately run firms. The five themes that emerged as self-discipline, workforce attitude challenges, ethical conduct, external institutional barriers, and practical strategies reflect a complex interplay between internal leadership behaviors and the broader socio-cultural and institutional environment in which these businesses operate. This study highlights how entrepreneurial engineers, as business owners, act as both technical leaders and organizational architects, shaping discipline not through rigid hierarchies but through adaptive, ethically anchored leadership. The findings contribute to a deeper understanding of engineering entrepreneurship in the Global South and offer practical implications for those leading technical enterprises in culturally complex and resource-constrained settings.

Malaria remained one of the most pressing public health challenges in sub-Saharan Africa, where over 90% of global cases and deaths occur annually. Timely diagnosis and effective treatment were essential for reducing mortality and curbing transmission, yet health system limitations often restrict access to formal healthcare in rural communities. Community health workers (CHWs) had been deployed widely to bridge this gap, offering diagnostic and treatment services for uncomplicated malaria in resource-constrained settings. This review evaluated the role of CHWs in malaria case management across rural sub-Saharan Africa, focusing on their effectiveness, challenges, and future opportunities. A structured literature review was conducted using PubMed, Scopus, and Web of Science databases to identify peer-reviewed publications from 2012 to 2025, with inclusion criteria emphasizing studies of CHW malaria diagnosis, treatment, supervision, and outcomes in African rural contexts. Evidence showed that CHW programs significantly increased access to prompt diagnosis through rapid diagnostic tests (RDTs) and ensure timely treatment with artemisinin-based combination therapies (ACTs), contributing to reductions in morbidity and mortality. However, challenges such as stockouts, inadequate training, supervision deficits, and limited integration with formal health systems constrained their impact. Innovative models integrating digital health tools, sustained supervision, and supply chain strengthening showed promise in addressing these barriers. CHWs represent a critical frontline strategy for malaria case management in rural Africa. Ensuring sustained investment, supportive supervision, and integration within national health systems is essential for optimizing their contribution to malaria elimination efforts. Keywords: Community health workers, Malaria, Sub-Saharan Africa, Case management, Rural health

Too sick to go home, too poor to stay: managing diabetic ketoacidosis in a resource-constrained setting

Background: Alzheimer’s disease (AD) is a leading worldwide cause of cognitive impairment, necessitating accurate, inexpensive diagnostic tools to enable early recognition. Methods: In this study, we present a robust deep learning approach for AD classification based on structural MRI scans, ConvNeXt, an emergent convolutional architecture inspired by vision transformers. We introduce AlzaSet, a clinically curated T1-weighted MRI dataset of 79 subjects (63 with Alzheimer’s disease [AD], 16 cognitively normal controls [NC]) acquired on a 1.5 T Siemens Aera in axial, coronal, and sagittal planes, respectively (12,947 slices in total). Images are neuroradiologist-labeled. Results are reported per plane, with awareness of the class imbalance at the subject level. We further present AlzaSet, a novel, expertly labeled clinical dataset with axial, coronal, and sagittal perspectives from AD and cognitively normal control subjects. Three ConvNeXt sizes (Tiny, Small, Base) were compared and benchmarked against existing state-of-the-art CNN models (VGG16, VGG19, InceptionV3, DenseNet121). Results: ConvNeXt-Base consistently outperformed the other models on coronal slices with an accuracy of 98.37% and an AUC of 0.992. Coronal views were determined to be most diagnostically informative, with emphasis on visualization of the medial temporal lobe. Moreover, comparison with recent ensemble-based techniques showed superior performance with comparable computational efficiency. Conclusions: These results indicate that ConvNeXt-capable models applied to clinically curated datasets have strong potential to provide scalable, real-time AD screening in diverse settings, including both high-resource and resource-constrained settings.