Browse Articles

Correction: Effects of nitrogen, phosphorus and potassium formula fertilization on the yield and berry quality of blueberry

Translation, adaptation and validation of an epilepsy screening instrument in two Ghanaian languages

Introduction The prevalence of epilepsy in sub-Saharan Africa varies considerably, and the exact estimate for Ghana remains unclear, particularly in peri-urban areas where data are scarce. More community-based studies are required to understand better the actual burden of epilepsy in these areas and the difficulties in accessing healthcare. Objective To adapt and validate a household survey epilepsy-screening instrument in Shai-Osudoku and Ningo-Prampram District of Greater Accra Region, Ghana. Methods We developed a 17-item epilepsy screening instrument by modifying previously validated English language questionnaires. We included questions that could identify convulsive and non-convulsive seizures. Language experts forward- and back-translated the questionnaires into the two languages: Asante Twi and Dangme. Cases were people with confirmed epilepsy attending healthcare facilities where these languages are used. Controls were unaffected relatives of cases or people attending the same healthcare facilities for other medical conditions. We matched cases and controls for geographical location and ethnicity. An affirmative response to one of the seventeen questions by a participant was deemed a positive screen. The questionnaires were divided into two stages. The first stage consisted of broader, more general questions aimed at identifying potential cases of epilepsy. The second stage involved a more detailed and focused set of questions administered to those who screened positive in the first stage. Results One hundred and forty Dangme speakers (70 cases and 70 controls) and 100 Asante Twi speakers (50 cases and 50 controls) were recruited. The sensitivity and specificity for Dangme were: Stage 1, 100% and 80%, and Stage 2, 98.6% and 85.7%. The Dangme version reliably identified epilepsy with positive predictive values of 83.3% and 87.3% at stages 1 and 2. The questionnaire excluded epilepsy with 100% and 98.4% negative predictive values. For the Asante Twi version, the sensitivity and specificity were 98% and 92% (95% at Stage 1, and for Stage 2, 96% and 94%. The Asante Twi questionnaire reliably specified epilepsy with positive predictive values of 92.5% and 94.1% at stages 1 and 2. It excluded epilepsy with negative predictive values of 97.9% and 95.9% for the two stages Conclusions Our questionnaire is valid for the two languages and usable for community-based epilepsy surveys in Ghana. It can also be adapted for other resource-poor settings, although translation and iterative in-country testing will be needed to ensure its validity.

Investigating the performance of multivariate LSTM models to predict the occurrence of Distributed Denial of Service (DDoS) attack

In the current cybersecurity landscape, Distributed Denial of Service (DDoS) attacks have become a prevalent form of cybercrime. These attacks are relatively easy to execute but can cause significant disruption and damage to targeted systems and networks. Generally, attackers perform it to make reprisal but sometimes this issue can be authentic also. In this paper basically conversed about some deep learning models that will hand over a descent accuracy in prediction of DDoS attacks. This study evaluates various models, including Vanilla LSTM, Stacked LSTM, Deep Neural Networks (DNN), and other machine learning models such as Random Forest, AdaBoost, and Gaussian Naive Bayes to determine the DDoS attack along with comparing these approaches as well as perceiving which one is about to give elegant outcomes in prediction. The rationale for selecting Long Short-Term Memory (LSTM) networks for evaluation in our study is based on their proven effectiveness in modeling sequential and time-series data, which are inherent characteristics of network traffic and cybersecurity data. Here, a benchmark dataset named CICDDoS2019 is used that contains 88 features from which a handful (22) convenient features are extracted further deep learning models are applied. The result that is acquired here is significantly better than available techniques those are attainable in this context by using Machine Learning models, data mining techniques and some IOT based approaches. It’s not possible to completely avoid your server from these threats but by applying discussed techniques in the present juncture, these attacks can be prevented to an extent and it will also help to server to fulfil the genuine requests instead of sticking in the accomplishing the requests created by the unauthentic user.

Woody species diversity, structure and community distribution along environmental gradients of Seqela Dry Afromontane forest in Northwestern Ethiopia

Dry evergreen Afromontane forests are severely threatened due to the expansion of agriculture and overgrazing by livestock. The objective of this study was to investigate the composition of woody species, structure, regeneration status and plant communities in Seqela forest, as well as the relationship between plant community types and environmental variables. Systematic sampling was used to collect vegetation and environmental data from 52 (20 m x 20 m) (400 m2) plots. Density, Diameter at Breast Height (DBH), basal area, frequency, and importance value index (IVI) of woody species were computed to characterize the vegetation structure of the forest. Agglomerative hierarchical cluster analysis and Canonical Correspondence Analysis (CCA) with R software were used to identify plant communities and analyse the relationship between plant community types and environmental variables, respectively. A total of 68 woody plant species belonging to 63 genera and 44 families were identified. The Shannon diversity index and evenness values of the study area were 2.12 and 0.92, respectively. The total basal area and density of woody species were 27.4 m2 ha−1 and 1079.3 individual ha−1, respectively. The most frequent woody species in the Seqela forest included Albizia gummifera (51.92%), Croton macrostachyus (44.23%), Olinia rochetiana and Teclea nobilis (36.54%). Additionally, the most dominant species, as indicated by their importance value index (IVI), were Erythrina brucei (IVI = 11.24), Prunus africana (IVI=8.68), and Croton macrostachyus (IVI=7.38). Four plant community types were identified: Albizia gummifera - Ekebergia capensis, Prunus africana - Croton macrostachyus, Vachellia abyssinica - Dombeya torrida and Schefflera abyssinica - Teclea nobilis. The CCA results showed that the variation of species distribution and plant community formation were significantly (P < 0.05) related to altitude, organic matter, aspect, slope and soil available phosphorus. The regeneration status assessment of the forest revealed a good regeneration status, which was linked to diverse and abundant seed bank in the soil can ensure a continuous supply of seeds for regeneration; therefore, it is recommended to implement periodic soil seed bank assessments to monitor seed diversity and abundance and inform targeted conservation actions.

Exploring medical error taxonomies and human factors in simulation-based healthcare education

This study aims to provide an updated overview of medical error taxonomies by building on a robust review conducted in 2011. It seeks to identify the key characteristics of the most suitable taxonomy for use in high-fidelity simulation-based postgraduate courses in Critical Care. While many taxonomies are available, none seem to be explicitly designed for the unique context of healthcare simulation-based education, in which errors are regarded as essential learning opportunities. Rather than creating a new classification system, this study proposes integrating existing taxonomies to enhance their applicability in simulation training. Through data from surveys of participants and tutors in postgraduate simulation-based courses, this study provides an exploratory analysis of whether a generic or domain-specific taxonomy is more suitable for healthcare education. While a generic classification may cover a broad spectrum of errors, a domain-specific approach could be more relatable and practical for healthcare professionals in a given domain, potentially improving error-reporting rates. Seven strong links were identified in the reviewed classification systems. These correlations allowed the authors to propose various simulation training strategies to address the errors identified in both the classification systems. This approach focuses on error management and fostering a safety culture, aiming to reduce communication-related errors by introducing the principles of Crisis Resource Management, effective communication methods, and overall teamwork improvement. The gathered data contributes to a better understanding and training of the most prevalent medical errors, with significant correlations found between different medical error taxonomies, suggesting that addressing one can positively impact others. The study highlights the importance of simulation-based education in healthcare for error management and analysis.

Toll-like receptor 2/6-stimulated HMC-1 mast cells promote keratinocyte migration in wound healing

Mast cells, immune sentinels that respond to various stimuli in barrier organs, provide defense by expressing pattern recognition receptors, such as Toll-like receptors (TLRs). They may affect inflammatory responses and wound healing. Here, we investigated the effect of TLR2/6-stimulated mast cells on wound healing in keratinocytes. The human mast cell line HMC-1 was treated with the TLR2/6 agonist FSL-1, and the conditioned medium (CM) was collected from untreated cells (HMC-1 CM) and FSL-1-stimulated cells (FSL-1-HMC-1 CM). Cell migration was evaluated in keratinocyte cells (HaCaT) treated with HMC-1 CM and FSL-1-HMC-1 CM via scratch and Transwell assays. Mice were treated with HMC-1 CM, FSL-1-HMC-1 CM, and FSL-1. Wound closure was measured, and tissue regeneration was assessed using hematoxylin and eosin staining. Growth factor expression levels were evaluated to identify the factors affecting wound healing. The tryptase inhibitor APC 366 was treated with HMC-1 CM and FSL-1-HMC-1 CM in a scratch assay. This study revealed that HMC-1 CM affected HaCaT cell migration, which was facilitated by FSL-1-HMC-1 CM. HMC-1 CM promoted tryptase-dependent HaCaT migration. Moreover, FSL-1-HMC-1 CM enhanced wound healing in C57BL/6J mice in vivo. Our findings indicated that TLR2/6-stimulated mast cells contributed to skin homeostasis by promoting tryptase-dependent wound healing.

Spontaneous oxycodone withdrawal disrupts sleep, diurnal, and electrophysiological dynamics in rats

Opioid dependence is defined by an aversive withdrawal syndrome upon drug cessation that can motivate continued drug-taking, development of opioid use disorder, and precipitate relapse. An understudied but common opioid withdrawal symptom is disrupted sleep, reported as both insomnia and daytime sleepiness. Despite the prevalence and severity of sleep disturbances during opioid withdrawal, there is a gap in our understanding of their interactions. The goal of this study was to establish an in-depth, temporal signature of spontaneous oxycodone withdrawal effects on the diurnal composition of discrete sleep stages and the dynamic spectral properties of the electroencephalogram (EEG) signal in male rats. We continuously recorded EEG and electromyography (EMG) signals for 8 d of spontaneous withdrawal after a 14-d escalating-dose oxycodone regimen (0.5–8.0 mg/kg, 2×d; SC). During withdrawal, there was a profound loss (peaking on days 2–3) and gradual return of diurnal structure in sleep, body temperature, and locomotor activity, as well as decreased sleep and wake bout durations dependent on lights on/off. Withdrawal was associated with significant alterations in the slope of the aperiodic 1/f component of the EEG power spectrum, an established biomarker of arousal level. Early in withdrawal, NREM exhibited an acute flattening and return to baseline of both low (1–4 Hz) and high (15–50 Hz) frequency components of the 1/f spectrum. These findings suggest temporally dependent withdrawal effects on sleep, reflecting the complex way in which the allostatic forces of opioid withdrawal impinge upon sleep and diurnal processes. These foundational data based on continuous tracking of vigilance state, sleep stage composition, and spectral EEG properties provide a detailed construct with which to form and test hypotheses on the mechanisms of opioid-sleep interactions.

Correction: Integration of evidence into Theory of Change frameworks in the healthcare sector: A rapid systematic review

The impact of straw and its post-pyrolysis incorporation on functional microbes and mineralization of organic carbon in yellow paddy soil

The impact of straw and biochar on carbon mineralization and the function of carbon cycle genes in paddy soil is important for soil nutrient management and the transformation of carbon pools. This research is based on a five-year field experiment with four treatments: no fertilizer application (CK); chemical fertilizer only (NPK); straw combined with chemical fertilizer (NPKS); and biochar combined with chemical fertilizer (NPKB). By integrating indoor mineralization culture with metagenomic approaches, we analyzed the response of organic carbon mineralization and carbon cycle genes in typical paddy soil from Guizhou Province, China, to different fertilization treatments. The result shows that the various fertilization treatments significantly increased the levels of soil organic carbon, dissolved organic carbon, microbial biomass carbon, and readily oxidizable organic carbon. The NPKS treatment increased the rate of soil organic carbon mineralization, whereas the NPKB treatment decreased it. Overall, the NPK and NPKB treatments increased the relative abundance of carbon fixation genes. The NPKS treatment increased the relative abundance of carbon degradation genes. The NPKS treatment increased the abundance of Proteobacteria, whereas the NPKB treatment decreased the abundance of Actinobacteria. Biochar after straw pyrolysis can reduce carbon loss and enhance sequestration of soil carbon, whereas straw decreases soil organic carbon stability, accelerating the transformation of soil carbon pools. Future research should encompass long-term impact assessments to comprehensively understand the enduring effects of these fertilization treatments on soil carbon mineralization and the function of carbon cycle genes.

Microscopic evidence for nanoparticle-mediated growth of native gold in sulfide deposits at the Higashi–Aogashima Knoll Caldera hydrothermal field

Gold (or electrum) in hydrothermal fluid precipitates directly from gold sulfide complex and/or partly via suspended nanoparticles. The hydrothermal fluid contains “invisible gold” that is atomically dispersed in sulfide minerals or as nanoparticles with a size of less than 10 nm. However, the contribution of these gold nanoparticles to the formation of native gold and its alloy with silver (electrum) remains unclear. The Higashi–Aogashima Knoll Caldera hydrothermal field, south of Tokyo, Japan, is an area of significant seafloor hydrothermal activity that is known for high-grade gold-containing minerals in sulfide-rich rocks. In this study, dry-polished thin sections were created to minimize sample damage and scanning and transmission electron microscopy were used to investigated the cross-sectional and three-dimensional morphologies of native gold grains in a sulfide-rich mound rock from the Central Cone site of the caldera. The surfaces of the gold grains comprised nanoparticles with sizes of 5–50 nm that were also attached to their periphery, which suggests that gold nanoparticles in deep-sea hydrothermal fluid were involved in the mineralization of the gold. In addition, the distribution of silver was uneven within the gold grains, which suggests that the gold precipitation comprised multiple stages at different temperatures that resulted in the post-deposition or secondary remobilization of silver.

LARP3 inhibits the apoptosis of hepatocellular carcinoma via the ROS/PI3K/c-Fos axis

Primary hepatocellular carcinoma (PHC) is the sixth most common cancer and the third leading cause of cancer death worldwide. Hepatocellular carcinoma (HCC) accounts for 75%-85% of PHC. LARP3 is aberrantly expressed in multiple cancers. We found that it is significantly highly expressed in the liver cancer tissues of HCC patients, but the exact role and specific mechanism of this abnormal expression are not yet clear. In this study, through bioinformatics analysis, we concluded that LARP3 expression is associated with a poor prognosis for patients with HCC. Through cellular experiments such as gene editing and phenotypic functions, we found that LARP3 promotes the occurrence and development of HCC and inhibits apoptosis. Finally, through biological means such as RNA sequencing, flow cytometry, western blotting, and the construction of a subcutaneous tumorigenesis model in nude mice, we concluded that inhibition of HCC apoptosis by LARP3 is related to LARP3 negatively regulating ROS level and inhibiting the PI3K/c-Fos/apoptosis axis. This study will provide potential targets for the treatment of HCC.

Classification-augmented survival estimation (CASE): A novel method for individualized long-term survival prediction with application to liver transplantation

Survival analysis is critical in many fields, particularly in healthcare where it can guide medical decisions. Conventional survival analysis methods like Kaplan-Meier and Cox proportional hazards models to generate survival curves indicating probability of survival v. time have limitations, especially for long-term prediction, due to assumptions that all instances follow a general population-level survival curve. Machine learning classification models, even those designed for survival predictions like random survival forest (RSF), also struggle to provide accurate long-term predictions due to class imbalance. We improve upon traditional survival machine learning approaches through a novel framework called classification-augmented survival estimation (CASE), which treats survival as a classification task that ultimately yields survival curves, beginning with dataset augmentation to improve class imbalance for use with any classification model. Unlike other approaches, CASE additionally provides an exact survival time prediction. We demonstrate CASE on a liver transplant case study to predict >20 years survival post-transplant, finding that CASE dataset augmentation improved AUCs from 0.69 to 0.88 and F1 scores from 0.32 to 0.73. Compared to Kaplan-Meier, Cox, and RSF survival models, the CASE framework demonstrated better performance across various existing survival metrics, as well as our novel metric, mean of individual areas under the survival curve (mAUSC). Further, we develop novel temporal feature importance methods to understand how different features may vary in survival importance over time, potentially providing actionable insights in real-world survival problems.

Passive earth pressure analysis considering hydraulic and mechanical hysteresis for unsaturated soil

This paper addresses the mechanical characteristics of a passive earth pressure problem taking into account water retention curve (SWRC) hysteresis. Both hydraulic (drying and wetting cycles) and mechanical hysteresis were considered. Parametric studies were carried out at various air entry values (AEV = 5–30 kPa), different wall frictions (δ = 0, 0.33 ϕ, 0.5 ϕ, 0.67 ϕ and δ = ϕ), and unsaturated conditions (covering the zone from the fully saturated to the transition suction, 0 - ~40 kPa) using an upper bound theorem. The numerical results were compared with a passive earth pressure equation based on the limit equilibrium method. The results indicated significant influences of SWRC hysteresis and wall frictions on the passive thrust (Pp) for the modeled suction profiles. An increase of 1.31-fold in Pp was obtained when the AEV changed from 10 to 20 kPa at the water table (Hw) = -2 m. Additionally, the combined effects of AEV and δ were found to be significant. An increase of 4.14-fold in Pp at Hw = -3 m was obtained when comparing the case of AEV = 30 kPa and δ = ϕ with AEV = 5 kPa and δ = 0. Based on the parametric studies, a series of design charts of the exerted passive thrust were proposed.

Economic impact and clinical benefits of clinical pharmacy interventions: A six-year multi-center study using an innovative medication management tool

Background Increasing healthcare costs, particularly in Low- and Middle-Income Countries (LMICs) like Egypt, highlight the need for rational economic strategies. Clinical pharmacy interventions offer potential benefits by reducing drug therapy problems and associated costs, thereby supporting healthcare system sustainability. Objective This study evaluates the economic impact and clinical benefits of clinical pharmacy interventions in four tertiary hospitals in Egypt by implementing an innovative tool for medication management, focusing on cost avoidance and return on investment (ROI), while accounting for case severity and drug therapy problem (DTP) resolution. Methods Utilizing a digital tool, cost avoidance was assessed by considering the severity of patient cases and the effectiveness of clinical pharmacists’ interventions in resolving identified DTPs. Additionally, ROI was calculated by incorporating both full-time equivalent (FTE) and non-FTE costs to evaluate the overall economic impact of clinical pharmacy services over a six-year period across four tertiary care hospitals in Egypt. Results Over six years, a total of 492,612 patients were reviewed, leading to 19,240 comprehensive clinical pharmacy interventions. These interventions achieved an 88.63% resolution of DTPs, significantly reducing patient risk by preventing adverse DTP consequences, resulting in a total cost avoidance of EGP 265.32 million (USD 8.60 million) and an average ROI of 7.6 (760%). This underscores the substantial economic impact of clinical pharmacy services, particularly in LMICs and countries transitioning to universal health insurance coverage, where cost efficiency and patient safety are critical. Conclusion This study underscores the importance of clinical pharmacy interventions in improving healthcare outcomes and generating significant economic benefits, particularly in low- and middle-income countries. By accounting for case severity and the level of DTP consequences, along with the efficiency of clinical pharmacist-led interventions in resolving DTPs, the economic impact of these services can be more accurately evaluated. These findings are essential for informing policy decisions, highlighting the critical role of clinical pharmacy services in supporting healthcare systems facing economic constraints.

Glaucoma detection and staging from visual field images using machine learning techniques

Purpose In this study, we investigated the performance of deep learning (DL) models to differentiate between normal and glaucomatous visual fields (VFs) and classify glaucoma from early to the advanced stage to observe if the DL model can stage glaucoma as Mills criteria using only the pattern deviation (PD) plots. The DL model results were compared with a machine learning (ML) classifier trained on conventional VF parameters. Methods A total of 265 PD plots and 265 numerical datasets of Humphrey 24–2 VF images were collected from 119 normal and 146 glaucomatous eyes to train the DL models to classify the images into four groups: normal, early glaucoma, moderate glaucoma, and advanced glaucoma. The two popular pre-trained DL models: ResNet18 and VGG16, were used to train the PD images using five-fold cross-validation (CV) and observed the performance using balanced, pre-augmented data (n = 476 images), imbalanced original data (n = 265) and feature extraction. The trained images were further investigated using the Grad-CAM visualization technique. Moreover, four ML models were trained from the global indices: mean deviation (MD), pattern standard deviation (PSD) and visual field index (VFI), using five-fold CV to compare the classification performance with the DL model’s result. Results The DL model, ResNet18 trained from balanced, pre-augmented PD images, achieved high accuracy in classifying the groups with an overall F1-score: 96.8%, precision: 97.0%, recall: 96.9%, and specificity: 99.0%. The highest F1 score was 87.8% for ResNet18 with the original dataset and 88.7% for VGG16 with feature extraction. The DL models successfully localized the affected VF loss in PD plots. Among the ML models, the random forest (RF) classifier performed best with an F1 score of 96%. Conclusion The DL model trained from PD plots was promising in differentiating normal and glaucomatous groups and performed similarly to conventional global indices. Hence, the evidence-based DL model trained from PD images demonstrated that the DL model could stage glaucoma using only PD plots like Mills criteria. This automated DL model will assist clinicians in precision glaucoma detection and progression management during extensive glaucoma screening.

Characteristics of ChatGPT users from Germany: Implications for the digital divide from web tracking data

A major challenge of our time is reducing disparities in access to and effective use of digital technologies, with recent discussions highlighting the role of AI in exacerbating the digital divide. We examine user characteristics that predict usage of the AI-powered conversational agent ChatGPT. We combine behavioral and survey data in a web tracked sample of N = 1376 German citizens to investigate differences in ChatGPT activity (usage, visits, and adoption) during the first 11 months from the launch of the service (November 30, 2022). Guided by a model of technology acceptance (UTAUT-2), we examine the role of socio-demographics commonly associated with the digital divide in ChatGPT activity and explore further socio-political attributes identified via stability selection in Lasso regressions. We confirm that lower age and higher education affect ChatGPT usage, but do not find that gender or income do. We find full-time employment and more children to be barriers to ChatGPT activity. Using a variety of social media was positively associated with ChatGPT activity. In terms of political variables, political knowledge and political self-efficacy as well as some political behaviors such as voting, debating political issues online and offline and political action online were all associated with ChatGPT activity, with online political debating and political self-efficacy negatively so. Finally, need for cognition and communication skills such as writing, attending meetings, or giving presentations, were also associated with ChatGPT engagement, though chairing/organizing meetings was negatively associated. Our research informs efforts to address digital disparities and promote digital literacy among underserved populations by presenting implications, recommendations, and discussions on ethical and social issues of our findings.

Characterizing the within-person variability of food insecurity in everyday life

Food insecurity (FI), the lack of access to adequate food, is linked with negative health and psychological outcomes. FI is typically measured retrospectively over the last year; although this measurement is useful to understand FI prevalence to inform broad policy, it leaves the experience of FI in everyday life poorly understood. Understanding how FI varies across shorter periods of time (days or weeks) can help inform FI prevention and/or intervention. This study characterizes within-person (day-to-day) variance in FI in everyday life. Low-income parents of school-aged children at risk for FI (n = 153) completed daily text message surveys in two-week bursts. Daily FI was measured with 4 yes/no items ranging in severity: worry about food, parent eating less than they should, child eating less than they should, and skipping meals. Items were analyzed as a sum score and individually (to examine FI severity). Among parents who reported FI at least once, FI meaningfully varied day-to-day within individuals (~26% of variation). Different indicators of FI, however, had different proportions of between- and within-person variability: Worry about food, a less severe aspect of FI, had 32% daily variation, whereas the more severe aspect of meal skipping had 45% daily variation. Thus, although substantial between-person differences in FI exist, there is meaningful within-person variability in FI. -person FI variability may be related to the indicator (e.g., severe FI shows greater within-person variability). Considering within-person FI variability, and not just average FI level, may help us understand how FI undermines functioning and how and when best to intervene.

Flexible and modular latent transition analysis—A tutorial using R

Latent transition analysis (LTA) is a useful statistical modelling approach for describe transitions between latent classes over time. LTA may be characterized in terms of prevalence at each time point and through transition probabilities over time. Investigating predictors of these transitions is often of key interest. Currently, LTA can mostly be carried out using commercial and specialized software and only to some limited extent by means of open source statistical software. This tutorial demonstrates a flexible and modular approach for LTA, providing a powerful alternative using R through a combination latent class analysis and multiple logistic regression models. This approach has several advantages from a modelling perspective, as demonstrated through revisiting a previously conducted LTA, published in PLoS ONE recently. In short, results were very similar to the original analysis using commercial software although some additional novel results were also obtained. The proposed alternative approach offers more options in terms of choice of effect measures, model assumptions such as hierarchical structures and covariate adjustment, and differential handling of missing data. R code snippets are provided in the tutorial. A detailed accompanying script is also provided for full reproducibility.

Comparison of mortality in people with type 2 diabetes between different ethnic groups: Systematic review and meta-analysis of longitudinal studies

Aims Type 2 diabetes (T2D) is more common in certain ethnic groups. This systematic review compares mortality risk between people with T2D from different ethnic groups and includes recent larger studies. Methods We searched nine databases using PRISMA guidelines (PROSPERO CRD42022372542). We included community-based prospective studies among adults with T2D from at least two different ethnicities. Two independent reviewers undertook screening, data extraction and quality assessment using the Newcastle-Ottawa Scale. The primary outcome compared all-cause mortality rates between ethnic groups (hazard ratio (HR) with 95% confidence intervals). Results From 30,825 searched records, we included 13 studies (7 meta-analysed), incorporating 573,173 T2D participants; 12 were good quality. Mortality risk was lower amongst people with T2D from South Asian [HR 0.68 (0.65–0.72)], Black [HR 0.82 (0.77–0.87)] and Chinese [HR 0.57 (0.46–0.70)] ethnicity compared to people of White ethnicity. Narrative synthesis corroborated these findings but demonstrated that people of indigenous Māori ethnicity had greater mortality risk compared to European ethnicity. Conclusions People with T2D of South Asian, Black and Chinese ethnicity have lower all-cause mortality risk than White ethnicity, with Māori ethnicity having higher mortality risk. Factors explaining mortality differences require further study, including understanding complication risk by ethnicity, to improve diabetes outcomes.

RETRACTED: The artificial intelligence-based agricultural field irrigation warning system using GA-BP neural network under smart agriculture

This work explores an intelligent field irrigation warning system based on the Enhanced Genetic Algorithm—Backpropagation Neural Network (EGA-BPNN) model in the context of smart agriculture. To achieve this, irrigation flow prediction in agricultural fields is chosen as the research topic. Firstly, the BPNN principles are studied, revealing issues such as sensitivity to initial values, susceptibility to local optima, and sample dependency. To address these problems, a genetic algorithm (GA) is adopted for optimizing the BPNN, and the EGA-BPNN model is used to predict irrigation flow in agricultural fields. Secondly, the EGA-BPNN model can overcome the local optimization and overfitting problems of traditional BPNN through the global search ability of GA. Moreover, it is suitable for the irrigation flow prediction task with complex environmental factors in smart agriculture. Finally, comparative experiments compare the prediction accuracy of BPNN and EGA-BPNN using single and dual water level flow prediction models respectively. The results reveal that as the number of nodes in the hidden layer increases, the model’s Mean Squared Error (MSE) and Relative Error (RE) show a decreasing trend, indicating an improvement in model prediction accuracy. When the number of nodes in the hidden layer increases from 6 to 16, the MSE of the single and dual water level flow prediction models decreases from 4.53×10 −4 to 3.68×10 −4 and 2.38×10 −4 to 1.66×10 −4 , respectively. Under a standalone BPNN, the absolute relative error in flow prediction is 1.09%. In contrast, the EGA-BPNN model achieves a significantly lower mean absolute relative error of 0.41% for single-flow prediction, demonstrating superior prediction performance. Furthermore, compared to the BPNN, the EGA-BPNN model exhibits a 2.11 reduction in MSE, further emphasizing the positive impact of introducing the GA on model performance. The research outcomes contribute to more accurate water resource planning and management, providing a more reliable basis for decision-making.