Explore the vast range of titles available.

Because biometric data is sensitive, centralized training poses a privacy risk, even though biometric recognition is essential for contemporary applications. Federated learning (FL), which permits decentralized training, provides a privacy-preserving substitute. Conventional FL, however, has trouble with interpretability and heterogeneous data (non-IID). In order to handle non-IID biometric data, this framework adds an attention mechanism at the central server that weights local model updates according to their significance. Differential privacy and secure update protocols safeguard data while preserving accuracy. The A3-FL framework is evaluated in this study using FVC2004 fingerprint data, with each client's features extracted using a Siamese Convolutional Neural Network (Siamese-CNN). By dynamically modifying client contributions, the attention mechanism increases the accuracy of the global model.The accuracy, convergence speed, and robustness of the A3-FL framework are superior to those of standard FL (FedAvg) and static baselines, according to experimental evaluations using fingerprint data (FVC2004). The accuracy of the attention-based approach was 0.8413, while FedAvg, Local-only, and Centralized approaches were 0.8164, 0.7664, and 0.7997, respectively. Accuracy stayed high at 0.8330 even with differential privacy. A scalable and privacy-sensitive biometric system for secure and effective recognition in dispersed environments is presented in this work.

Malaria remains a major public health concern in Ethiopia, particularly in the Amhara Region, where seasonal and unpredictable transmission patterns make prevention and control challenging. Accurately forecasting malaria outbreaks is essential for effective resource allocation and timely interventions. This study proposes a hybrid predictive modeling framework that combines time-series forecasting, multi-output regression, and conventional regression-based prediction to forecast the incidence of malaria. Environmental variables, past malaria case data, and demographic information from Amhara Region health centers were used to train and validate the models. The multi-output regression approach enables the simultaneous prediction of multiple outcomes, including Plasmodium species-specific cases, temporal trends, and spatial variations, whereas the hybrid framework captures both seasonal patterns and correlations among predictors. The proposed model exhibits higher prediction accuracy than single-method approaches, exposing hidden patterns and providing valuable information to public health authorities. This study provides a valid and repeatable malaria incidence prediction framework that can support evidence-based decision-making, targeted interventions, and resource optimization in endemic areas.

Background Morphological plasticity is one of the capacities of plants to modify their morphological appearance in response to external stimuli. A plant’s morphology and physiology are constantly tuned to its variable surroundings by complex interactions between environmental stimuli and internal signals. In most of plant species,, such phenotypic and physiological expression varies among different varieties based on their levels of particular environmental stress conditions. However, the morphological and yield responses of common bean varieties to different environmental conditions are not well known. The purpose of the study was to evaluate morphological and yield response of common bean to soil moisture stress and to investigate the morphological mechanism by which common bean varieties tolerate fluctuations in moisture stress. Methods A pot experiment was carried out to investigate the effects of different moisture levels on the phenotypic and yield responses of common bean varieties. A factorial combination of five common bean varieties (Hirna, kufanzik, Awash-1, Ado, and Chercher) and three moisture levels (control, waterlogging stress, and moisture deficit stress) was used in three replications. Moisture stress treatments were started 20 days after planting, at the trifoliate growth stage. To evaluate the response of each variety, morphological and yield data were collected at week intervals. Main results The results indicated that moisture levels and varieties had a significant influence on all growth parameters. Crop phenology was significantly influenced by the interaction effect of moisture level and variety. Exposing Hirna variety to moisture stress led to extended flowering and pod setting by 23 and 24 days, respectively, compared to the other treatments. The results showed that the phenotypic responses to moisture deficit and waterlogging stress varied between varieties. Waterlogging stress had a stronger reduction effect on the fresh weight, dry weight and leaf area of common bean varieties than moisture deficit and the control. Pods per plant, seeds per plant, grain yield per plant, and harvest index were significantly influenced by the varieties, moisture stress levels and their interaction. Except for Chercher and Hirna. However, varieties Ado, kufanzik and Awasha-1 did not show significant differences on the time of flower initiation due to moisture level. Biomass and growth in leaf fresh weight, leaf dry weight, leaf area, leaf number and plant height were significantly influenced by moisture level. When moisture deficit and waterlogging stress occurred, Ado and Awash-1 were more responsive to moisture stress than Hirna, Chercher, and Kufanzik. Conclusion Hence, Hirna and Kufanzik varieties were found to be tolerant because they produced higher yields than the Chercher, Awash-1, and Ado varieties.

Background Induced mutation through physical mutagens, such as gamma irradiation is an effective and complementary breeding tool to enhance crop improvement by increasing genetic variation and creating heritable mutant alleles. This is highly important in crops such as sweetpotato, where genetic and reproductive constraints limit conventional improvements. This study was therefore designed to investigate the effects of gamma irradiation on the growth, stomatal behavior, and photosynthetic pigments of sweet potato ( Ipomoea batatas L .) in the M1V1 generation. The experiment was conducted in a screenhouse in a randomized complete block design (RCBD) with twenty replications. Three sweet potato genotypes, Awassa-83, Alamura, and Kabode, were subjected to gamma irradiation doses of 0, 15, 25, and 35 Gray (Gy) to assess their median lethal dose (LD50), which was estimated via a probit model based on the seedling mortality rate, growth, stomatal behavior, and photosynthetic pigment responses. Results Significant differences ( P  < 0.05) were observed across varieties, doses, and their interactions for vine number, leaf number, internode length, vine length, and petiole length. The vine number was consistently reduced across the three varieties as the gamma dose increased to the maximum. The decline was 4.85 vines at 15 Gy in Alamura to Kabode’s 1.63 vines at 35 Gy. Similarly, the internode and vine length decreased with increasing dose across the varieties where Alamura scored the highest in both trait at 15 Gy but up to 30% decline in internode and 65% decline in vine length at the maximum dose of Kabode and Awassa-83, respectively. Stomatal traits response, including stomatal length and number, varied among the varieties with differing gamma dose following a non-linear pattern. Pigment analysis revealed that chlorophyll a, b and total chlorophyll was peaked in Awassa-83 and Alamura at 15 Gy before collapsing below 2 µg/ml at 35 Gy across all varieties. Carotenoids were moderately enhanced at 25 Gy in Awassa-83 (5.32 µg/ml) but declined sharply at 35 Gy. Conclusion These findings highlight the potential of gamma irradiation in creating variation among the varieties for important traits. Further, the varieties responded differently to the gamma irradiation doses, showing genotypic differences. This helps to identify potential mutants in the subsequent generation for the traits studied and their physiological ramifications.

Background Neonatal mortality is one of the major public health problems throughout the world and most notably in developing countries. There exist inconclusive findings on the effect of antenatal care visits on neonatal death worldwide. Thus, the aim of this systematic review and meta-analysis was to reveal the pooled effect of antenatal care visits on neonatal death. Methods The present systematic review and meta-analysis was performed using published literature, which was accessed from national and international databases such as, Medline/PubMed, EMBASE, CINAHL, Cochrane Central library, Google Scholar, and HINARI. STATA/SE for windows version 13 software was used to calculate the pooled effect size with 95% confidence intervals (95% CI) of maternal antenatal care visits on neonatal death using the DerSimonian and Laird random effects meta-analysis (random effects model), and results were displayed using forest plot. Statistical heterogeneity was checked using the Cochran Q test (chi-squared statistic) and I 2 test statistic and by visual examination of the forest plot. Results A total of 18 studies, which fulfilled the inclusion criteria, were included in the present systematic review and meta-analysis. The finding of the present systematic review and meta-analysis revealed that antenatal care visits decrease the risk of neonatal mortality [pooled effect size 0.66 (95% CI, 0.54, 0.80)]. Cochrane Q test ( P  < 0.001) revealed no significant heterogeneity among included studies, but I 2 statistic revealed sizeable heterogeneity up to 80.5% ( I 2  = 80.5%). In the present meta-analysis traditional funnel plot, Egger’s weighted regression ( P  = 0.48) as well as Begg’s rank correlation statistic ( P  = 0.47) revealed no evidence of publication bias. Conclusions The present systematic review and meta-analysis revealed that antenatal care visits were significantly associated with lower rates of neonatal death. The risk of neonatal death was significantly reduced by 34% among newborns delivered from mothers who had antenatal care visits. Thus, visiting antenatal care clinics during pregnancy is strongly recommended especially in resource-limited settings like countries of sub-Saharan Africa.

This review synthesizes global research on allometric models for estimating bamboo biomass across a wide range of species and ecological regions. A systematic search of four major scientific databases Scopus, Web of Science, Science Direct, and Google Scholar covering the period 2000–2025 identified 55 peer-reviewed studies that met defined inclusion criteria. The review evaluates the effectiveness, limitations, and applications of these models in supporting forest management, carbon sequestration, sustainable agriculture, and bioenergy production. Representative case studies from Asia, Africa, Latin America, and other regions reveal key methodological trends, including species-specific modeling, regional adaptation, and the use of standardized biometric parameters. Persistent challenges include limited data availability, restricted model transferability across regions, and the influence of structural variation among bamboo species on model accuracy. Recent innovations highlight the integration of remote sensing, LiDAR (Light Detection and Ranging), machine learning, and GIS (Geographic Information Systems) to improve model precision, scalability, and operational efficiency. The review underscores the importance of regionally calibrated models and proposes a hybrid framework that combines field-based measurements with advanced analytical tools to capture spatial and temporal variability in bamboo biomass. Finally, future research directions are outlined, focusing on enhancing model robustness, expanding geographic and taxonomic coverage, and improving policy relevance in the context of climate change mitigation and sustainable land-use planning.

Background Tuberculosis and diabetes mellitus are significant global public health challenges. In Sub-Saharan Africa, study findings regarding prevalence of diabetes mellitus amongst tuberculosis patients have been inconsistent and highly variable. Therefore, this systematic review and meta-analysis estimates the overall prevalence of diabetes mellitus among tuberculosis patients in Sub-Saharan Africa. Methods Four international databases (PubMed, Google Scholar, Science Direct and Cochrane Library) were systematically searched. We included all observational studies reporting the prevalence of DM among TB patients in Sub-Saharan Africa. All necessary data for this review were extracted using a standardized data extraction format by two authors (CT and AA1). STATA Version 14 statistical software was employed to conduct meta-analysis. The Cochrane Q test statistics and I 2 test were used to assess the heterogeneity of the studies. Finally, a random effects meta-analysis model was computed to estimate the pooled prevalence of diabetes mellitus in TB patients. Besides, subgroup analysis was done based on different factors. Results In the meta-analysis, sixteen studies fulfilled the inclusion criteria and were included. The findings of these 16 studies revealed that the pooled prevalence of diabetes mellitus among tuberculosis patients in Sub-Saharan Africa was 9.0% (95% CI: 6.0, 12.0%). The highest prevalence of diabetes mellitus among tuberculosis patients was found in Nigeria (15%), followed by Tanzania (11%), and then Ethiopia (10%). Besides, the prevalence of diabetes mellitus among HIV infected TB patients was (8.9%) which is slightly higher than HIV uninfected (7.7%) TB patients. Conclusion Diabetes mellitus among tuberculosis patients in Sub-Saharan Africa was significantly high. Moreover, this study found that there was a high prevalence of DM among HIV infected than uninfected TB patients. It is strongly recommended to screen for DM among TB patients and special emphasis should be given for early screening of DM among TB/HIV co-infected patients.

Infrastructure cracks are critical indicators of structural deterioration in pavements, bridges, and buildings. Automated crack segmentation has therefore become an important component of structural health monitoring systems. However, accurate pixel-level crack segmentation on resource-constrained devices remains challenging due to the thin, low-contrast, and curvilinear morphology of cracks, as well as severe foreground–background class imbalance. To address these challenges, we propose LiteCrackSeg, a lightweight hybrid CNN–transformer architecture designed for efficient and accurate crack segmentation. The proposed framework adopts a hybrid MobileViT encoder that captures both local spatial details and long-range contextual dependencies while maintaining a compact model size. To enhance morphological sensitivity to elongated crack structures, we introduce a Morphology-Aware MobileViT (MAM-ViT) bottleneck, which integrates dual-branch Dynamic Snake Convolutions (DSConv) to align receptive fields with crack trajectories. Furthermore, a transformer-based decoder with local self-attention progressively reconstructs spatial details, while an attention-guided multi-scale fusion strategy improves boundary precision and structural continuity. To mitigate severe class imbalance, the model is trained using the Tversky loss, which explicitly balances false positives and false negatives. Extensive experiments on three public crack segmentation datasets (DeepCrack, CrackMap, and TUT) demonstrate that LiteCrackSeg achieves state-of-the-art segmentation performance while maintaining high computational efficiency. The proposed model requires only 2.72M parameters and 3.23 GFLOPs, achieving real-time inference at 56 FPS on 512 × 512 images, making it suitable for deployment on resource-constrained edge devices for practical infrastructure inspection.

Background Food safety and hygiene are currently a global health concern, especially in unindustrialized countries, as a result of increasing food-borne diseases (FBDs) and accompanying deaths. It has continued to be a critical problem for people, food companies, and food control officials in developed and developing nations. Objective The objective of the study was to assess food safety practices and associated factors among food handlers in food establishments in Mettu and Bedelle towns, south-west Ethiopia, 2022. Methods A community-based cross-sectional study was conducted from February to March 2022, among 450 randomly selected food handlers working in food and drink establishments in Mettu and Bedelle towns, Southwest Ethiopia. Data was collected using an interviewer-administered structured questionnaire. The data was coded and entered into Epi Data version 3.1 before being exported to SPSS version 20 for analysis. Both bivariate and multivariable logistic regression models were fitted. An adjusted odds ratio and a 95% confidence level were estimated to assess the significance of associations. A p -value of < 0.05 was considered sufficient to declare the statistical significance of variables in the final model. Result A total of 450 food handlers participated in the study, making the response rate 99.3%. About 202 (44.9%) of respondents had poor practices in food safety. Lack of supervision (AOR = 6.2, 95% CI: 3.37, 11.39), absence of regular medical checkups (AOR = 1.98; 95% CI: 1.14, 3.43), lack of knowledge of food safety practices (AOR =2.32; 95% CI: 1.38, 3.89), availability of water storage equipment (AOR =0.37; CI: 0.21, 0.64), and unavailability of a refrigerator (AOR =0.24; 95% CI: 0.12) were factors significantly associated with food safety practices. Conclusion The level of poor food safety practices was remarkably high. Knowledge of food safety, medical checkups, service year as food handler, availability of water storage equipment, availability of refrigerator, and sanitary supervision were all significantly associated with food safety practice. Hence, great efforts are needed to improve food safety practices, and awareness should be created for food handlers on food safety.

Although Ethiopia has substantial improvements in various health indicators such as maternal and child mortality, the burden of neonatal mortality remains high. Between 2016 and 2019, neonatal mortality increased from 29 deaths per 1,000 live births to 33 deaths per 1,000 live births. This study aimed to explore the spatial patterns and factors contributing to neonatal mortality in Ethiopia. Data from the Ethiopian Demographic and Health Surveys (EDHS) for the years 2000, 2005, 2011, 2016, and 2019 were analyzed. The EDHS sampling design uses a two-stage cluster sampling technique, considering census enumeration areas as primary sampling units and households as secondary sampling units. We used the Spatial Scan analysis in SaTScan and Getis-Ord Gi* statistic in Geographic Information System (GIS), to analyse the spatiotemporal patterns of neonatal mortality. Maternal, newborn and health service-related factors contributing to neonatal mortality were also analyzed using a multilevel logistic regression model. Adjusted Odds Rios (AOR) with corresponding 95% CI were presented as a measure of association and a P-value of 0.05 was used to declare statistical significance. During the initial three consecutive surveys, there was a consistent pattern of hot spot clusters in the Amhara and Benshangul Gumuz regions, along with certain parts of the Oromia region. However, in later surveys, these clusters shifted to the eastern parts of the country, notably including the Somali region. Early initiation of breast feeding was associated with reduced chances of neonatal death (Adjusted Odds Ratio [AOR]) = 0.27; 95% Confidence Interval [CI]: 0.23, 0.32). Neonates born at home (AOR = 1.46; 95% CI: 1.16, 1.82) and male babies had a higher likelihood of mortality during the neonatal period compared to their counterparts (AOR = 1.36; 95% CI: 1.24, 1.51). The odds of neonatal mortality increased with the number of children a mother had ever given birth to (AOR = 1.36; 95% CI: 1.24, 1.51). In contrast, longer birth intervals were associated with a reduced risk of neonatal mortality (AOR = 0.76; 95% CI: 0.68, 0.83). The central southern, central-western, north-western, and northern parts of Ethiopia had most of the neonatal death clusters in the first three rounds of DHS while eastern Ethiopia had the highest neonatal mortality clusters in the latest two surveys. Our results underscore the importance for policymakers and health administrators to reassess intervention approaches and reallocate resources to regions identified as hot spots for neonatal mortality. Enhancing the initiation of breastfeeding within the first hour of birth would improve newborn survival rates. Special attention and care need to be given to babies born of smaller sizes.