Jatiya Kabi Kazi Nazrul Islam University

Maternal health is one of the pivotal concerns in rural Bangladesh, where many women still face obstacles to safe and timely care during pregnancy and childbirth. This study explores maternal health knowledge and their healthcare-seeking behavior during pregnancy, delivery, and after childbirth among the rural married women in Bangladesh. This is a qualitative data-based study, and it conducted 12 In-depth Interviews (IDIs) and 3 Key Informant Interviews (KIIs) in Salimpur village, Trishal municipality of Mymensingh District, Bangladesh. This study interviewed 12 married women of reproductive age (18–35 years) and 3 key informants (two household heads and a village doctor) using purposive sampling. In addition, the data were analyzed using reflexive thematic analysis, combining inductive and deductive approaches guided by the Social-Ecological Model. All interviews were audio-recorded, transcribed verbatim, and coded to ensure credibility. In this study four key themes emerged in the light of Social-Ecological Model, namely: (1) education as an enabler of health literacy and autonomy; (2) family gatekeeping through the decision-making authority of mothers-in-law and husbands; (3) emerging skepticism toward some restrictive cultural practices, particularly among younger and more educated women; and (4) quality preference over cost that drives the utilization of private healthcare despite financial difficulties or constraints. Rural women face multiple barriers to maternal healthcare services. For this, improving maternal health outcomes requires ensuring accessible and affordable services, and needs to improve women’s health knowledge and decision-making power. Moreover, findings of this study are contextually situated within one rural village and are not intended to be statistically generalizable, but offer transferable insights for rural Bangladesh and similar low-resource settings.

Childhood undernutrition remains a significant public health challenge in Bangladesh, affecting both the physical and cognitive development of children and the country’s economic growth. The study examined the spatial distribution and multilevel determinants of stunting, wasting, and underweight among children under five years of age in Bangladesh. Data were obtained from the nationally representative Bangladesh Demographic and Health Survey (BDHS) 2022. Spatial patterns were analyzed using Moran’s I to assess spatial autocorrelation and the Getis-Ord Gi* statistic to identify hot and cold spots. Three-level Bayesian multilevel logistic regression models were applied to determine the predictors of stunting, wasting, and underweight at the individual, household, and community levels. Overall, 22.5

Perovskite solar cells (PSCs) based on lead have achieved a record power conversion efficiency (PCE) in recent years; however, the toxicity and stability of Pb, especially of CH3NH3PbI3, limits its commercialization. As a result, much effort is being put into developing lead-free PSCs. This paper compares the performances of different types of lead-free PSCs and their choice of absorber materials using Solar Cell Capacitance Simulator-1D simulations. In this context, the use of MASnI3, MASnBr3, and MABiI3 as solvent-processed absorbers and their effects on Voc, Jsc, fill factor, and PCE are discussed in this part of the study snippets series. MASnI3 shows the best comprehensive cell efficiency with a maximum PCE of 30.04%. Similarly, MASnBr3 and MABiI3 deliver high performance, with maximum PCEs of 29.19% and 27.97%, respectively, obtained at optimized absorber thicknesses. The outcomes demonstrate the effect of the thickness of the absorber layer, the doping concentration, and the defect density on the performance of the device, revealing that the device with MASnI3 has proved superior performance under the above parameters. Additionally, it elaborates on the enhancement strategies for the hole transport layer and electron transport layer, propagating the idea that defect density at interfaces must be minimized. This study contributes to the ongoing efforts to make highly efficient, lead-free PSCs a practical reality while aiding in sustainable energy conversion and storage.

Energy demand in today's world is increasing. So, to meet this energy demand and reduce reliance on fossil fuels, we need to look to renewable energy sources, and hydrogen storage can be one of them. This study conducted a magnificent investigation into the thermo-dynamical stability, optoelectronic properties, hydrogen storage capability, and photocatalytic activity of Li2XBH6 (X = Li, Na, K) Double perovskite hydrides (DPH) using a density functional theory (DFT)-based approach, with ab initio molecular dynamics (AIMD) simulations, which specifies that the compounds are thermally stable. The electronic structure analysis indicated that all compounds exhibit semiconductor behavior, with band gap energies of 1.79 eV for Li2LiBH6, 1.66 eV for Li2NaBH6, and 1.48 eV for Li2KBH6, suggesting their potential for efficient photonic and optoelectronic applications such as solar cell absorbers. This study shows a prominent initial theoretical GHSC, with Li2LiBH6 demonstrating the highest storage potential at 16.05 wt%, Li2KBH6 the lowest at 8.66 wt%, and Li2NaBH6 11.25 wt% - all considerably surpassing the US-DOE target of >= 5.5 wt%. The study shows that these substances are promising theoretical candidates for advanced optoelectronic devices, hydrogen storage materials, and photocatalytic applications.

Diabetic Retinopathy (DR) is a leading cause of preventable blindness, and it is important to accurately detect it as early as possible, as well as to measure the severity to provide early clinical indicators. Fundus image assessment in the manual mode is subjective and labor-intensive, and is hard to scale, which has encouraged automated approaches. Current deep learning methods are usually based on either convolutional neural networks (CNNs) or transformer-based ones, focusing on local lesion features or global retina context separately. In addition, relational dependencies among lesions that are clinically significant in the case of severity development are under-modeled. This paper introduces a relational, hybrid deep learning model that integrates convolutional, transformer-based, and graph attention models to classify binary and multiclass DR. A ResNet-Graph Attention Network (ResNet-GAT) is introduced to explicitly model spatial relational dependencies among regional feature descriptors and iscompared with independent stand-alone ResNet-50 and Swin Transformer models. Also, a hybrid ResNet-Swin Transformer model combines fine-grained lesion representations and hierarchical global context. Experiments with the APTOS 2019 data show that the hybrid model has an accuracy of 98.09% with a Quadratic Weighted Kappa (QWK) of 0.9618 when used to classify binary, and 94.89% with a QWK of 0.9726 when used to classify five classes of severity. Robustness is proven by cross-dataset inference on IDRiD. The interpretability analysis performed through Grad-CAM shows that the predictions are made based on clinically significant areas, such as microaneurysms, hemorrhages, and exudates. It is demonstrated that performance and interpretability are enhanced by a combination of local, global, and relational representations in automated screening of DR.