Explore the vast range of titles available.

Purpose One of the United Nations’ sustainable development goals is to ensure the availability of improved drinking water for everyone. In this study, we examine the association between access to improved drinking water at the district level and child nutritional outcomes in Pakistan.Design/methodology/approach We employ district-level unbalanced panel data from Pakistan from various rounds of Pakistan Social and Living Standards Measurement Surveys and Multiple Indicators Cluster Surveys compiled by the Data4Pakistan, Pakistan District Development Portal. We examine the impact of the percentage of the population in a given district with access to clean drinking water on the percentage of stunted, underweight and wasted children in the district. The analysis proceeds in two steps. In the first step, we explore the spatial distribution of improved drinking water coverage and child development outcomes across districts. In the second step, we study their relationship by employing standard panel estimation methods and controlling for district characteristics.Findings The spatial analysis reveals the large disparity among districts and provinces in terms of improved drinking water coverage and child nutrition. The estimation results indicate that there is a significant association between the accessibility of improved drinking water and child development outcomes. The effect is significant for child stunting and underweight but not for child wasting. The impact appears to be stronger in rural districts. These findings are robust to alternate empirical strategies.Originality/value This is the first such study to examine the provision of improved drinking water at the district level in relation to child developmental outcomes in a developing country context.Peer reviewThe peer review history for this article is available at: https://publons.com/publon/10.1108/IJSE-09-2023-0739

Melanoma is a fatal type of skin cancer; the fury spread results in a high fatality rate when the malignancy is not treated at an initial stage. The patients’ lives can be saved by accurately detecting skin cancer at an initial stage. A quick and precise diagnosis might help increase the patient’s survival rate. It necessitates the development of a computer-assisted diagnostic support system. This research proposes a novel deep transfer learning model for melanoma classification using MobileNetV2. The MobileNetV2 is a deep convolutional neural network that classifies the sample skin lesions as malignant or benign. The performance of the proposed deep learning model is evaluated using the ISIC 2020 dataset. The dataset contains less than 2% malignant samples, raising the class imbalance. Various data augmentation techniques were applied to tackle the class imbalance issue and add diversity to the dataset. The experimental results demonstrate that the proposed deep learning technique outperforms state-of-the-art deep learning techniques in terms of accuracy and computational cost.

The diagnosis of leukemia is based on the simultaneous analysis of morphological patterns of hematological images and the presence of clinical indicators in written reports. Majority of machine learning models are unimodal and centralized. They are not able to integrate information with the institutions or give clinically useful explanations. This paper suggests a federated multimodal architecture that integrates Vision Transformers (ViT) and ClinicalBERT to encode images and classify texts to conduct joint leukemia diagnosis and staging in decentralized medical devices, respectively. Both modalities are synthesised into a single semantic space to form a cross-modal fusion layer, and binary diagnosis and multiclass staging are facilitated by dual output heads. The framework uses federated learning protocol which maintains the privacy of data by the fact that the local data does not move out of institutional boundaries. To improve the level of transparency, SHAP-based explanations are provided on each prediction, where both visual regions and clinical tokens are considered important. The results of the experiments indicate that the suggested system is more accurate and has a higher F1-score than unimodal and centralized baselines and also has interpretable and patient-specific explanation, which is consistent with clinical expectations. The architecture is robust in the non-IID data distributions and is scaled through simulated healthcare networks, which makes it appropriate to deploy to actual health care in diagnostic oncology.

Autism spectrum disorder (ASD) is a complex psychological syndrome characterized by persistent difficulties in social interaction, restricted behaviours, speech, and nonverbal communication. The impacts of this disorder and the severity of symptoms vary from person to person. In most cases, symptoms of ASD appear at the age of 2 to 5 and continue throughout adolescence and into adulthood. While this disorder cannot be cured completely, studies have shown that early detection of this syndrome can assist in maintaining the behavioural and psychological development of children. Experts are currently studying various machine learning methods, particularly convolutional neural networks, to expedite the screening process. Convolutional neural networks are considered promising frameworks for the diagnosis of ASD. This study employs different pre‐trained convolutional neural networks such as ResNet34, ResNet50, AlexNet, MobileNetV2, VGG16, and VGG19 to diagnose ASD and compared their performance. Transfer learning was applied to every model included in the study to achieve higher results than the initial models. The proposed ResNet50 model achieved the highest accuracy, 92%, compared to other transfer learning models. The proposed method also outperformed the state‐of‐the‐art models in terms of accuracy and computational cost. Experts are currently studying various machine learning methods, particularly convolutional neural networks, to expedite the screening process. Convolutional neural networks are considered promising frameworks for the diagnosis of ASD. This study employs different pre‐trained convolutional neural networks such as ResNet34, ResNet50, AlexNet, MobileNetV2, VGG16, and VGG19 to diagnose ASD and compared their performance. The proposed ResNet50 model achieved the highest accuracy, 92%, compared to other transfer learning models.

Alzheimer’s disease (AD) is a brain illness that causes gradual memory loss. AD has no treatment and cannot be cured, so early detection is critical. Various AD diagnosis approaches are used in this regard, but Magnetic Resonance Imaging (MRI) provides the most helpful neuroimaging tool for detecting AD. In this paper, we employ a DenseNet-201 based transfer learning technique for diagnosing different Alzheimer’s stages as Non-Demented (ND), Moderate Demented (MOD), Mild Demented (MD), Very Mild Demented (VMD), and Severe Demented (SD). The suggested method for a dataset of MRI scans for Alzheimer’s disease is divided into five classes. Data augmentation methods were used to expand the size of the dataset and increase DenseNet-201’s accuracy. It was found that the proposed strategy provides a very high classification accuracy. This practical and reliable model delivers a success rate of 98.24%. The findings of the experiments demonstrate that the suggested deep learning approach is more accurate and performs well compared to existing techniques and state-of-the-art methods.

Catheter ablation (CA) and anti-arrhythmic drugs (AADs) minimize implanted cardioverter-defibrillator (ICD) shocks in individuals with ischemic cardiomyopathy and an ICD, while the best strategy is still unknown. CA has been proposed as a potentially effective means of reducing the occurrence of ICD events in a number of studies; however, there were insufficient relevant dates from randomized controlled trials. A meta-analysis and systematic review of randomized controlled trials were carried out to evaluate the efficacy of CA for the prevention of VA in patients with ischemic heart disease. Cardiovascular mortality, an unscheduled hospitalization due to increasing heart failure, appropriate ICD shock, or serious treatment-related consequences comprised the composite primary outcome. AADs were examined in six trials (n = 1564; follow-up = 15 ± 8 months), while CA was evaluated in four trials (n = 682; follow-up = 12 ± 6 months). Both CA (odds ratio (OR) 0.65, 95% confidence interval (CI) 0.47-0.82, p = 0.001) and AADs (OR 0.76, 95% CI 0.32-0.84, p = 0.034) significantly reduced the number of suitable ICD interventions, with no discernible difference between the two treatment approaches. AADs were observed to reduce incorrect ICD interventions (OR 0.38, p = 0.001), but CA did not. During follow-up, there was no correlation seen between reduced mortality and either CA or AAD. When compared to AAD, CA decreased the composite endpoint of cardiovascular death, adequate ICD shock, heart failure-related hospitalization, or severe treatment-related consequences in ICD patients with ischemic cardiomyopathy and symptomatic VT.

Neural machine translation (NMT) has advanced with deep learning and large‐scale multilingual models, yet translating low‐resource languages often lacks sufficient training data and leads to hallucinations. This often results in translated content that diverges significantly from the source text. This research proposes a refined Contrastive Decoding (CD) algorithm that dynamically adjusts weights of log probabilities from strong expert and weak amateur models to mitigate hallucinations in low‐resource NMT and improve translation quality. Advanced large language NMT models, including ChatGLM and LLaMA, are fine‐tuned and implemented for their superior contextual understanding and cross‐lingual capabilities. The refined CD algorithm evaluates multiple candidate translations using BLEU score, semantic similarity, and Named Entity Recognition accuracy. Extensive experimental results show substantial improvements in translation quality and a significant reduction in hallucination rates. Fine‐tuned models achieve higher evaluation metrics compared to baseline models and state‐of‐the‐art models. An ablation study confirms the contributions of each methodological component and highlights the effectiveness of the refined CD algorithm and advanced models in mitigating hallucinations. Notably, the refined methodology increased the BLEU score by approximately 30% compared to baseline models.

Degenerative lumbar spine diseases, including disk herniation, spinal stenosis, and facet arthropathy, are leading causes of chronic lower back pain, significantly affecting patients worldwide. Diagnosing these conditions through MRI imaging is time-consuming and subjective. This study introduces a novel deep-learning model using the MobileNetV2-UNet architecture to automate the classification and segmentation of lumbar spine degenerative diseases. The MobileNetV2-UNet model integrates MobileNetV2 for efficient feature extraction and UNet for accurate segmentation. The model was trained and validated on the RSNA 2024 Lumbar Spine Degenerative Classification dataset, which includes labeled MRI images. Depthwise separable convolutions enhanced computational efficiency, while skip connections in the UNet architecture preserved spatial details. The combined Cross-Entropy and IoU loss functions optimized classification accuracy and segmentation quality. The model’s performance was evaluated using 5-fold cross-validation. The MobileNetV2-UNet model achieved a 94.93% accuracy, a 94.61% Dice Coefficient, and a 95.65% F1 score in classifying and segmenting various lumbar spine conditions, such as neural foraminal narrowing and spinal canal stenosis. The model’s efficient architecture enables real-time application in clinical settings while maintaining high performance across diverse datasets. The proposed MobileNetV2-UNet model offers a reliable, efficient solution for automating lumbar spine degenerative disease diagnosis, reducing manual workload, and improving diagnostic precision. Future work will focus on enhancing generalization, improving model interpretability, and applying the framework to other musculoskeletal and neurological conditions, making it a promising tool for clinical use. Graphical Abstract

Hate speech (HS) has grown because of increasing social media platform usage, which includes Twitter, YouTube, and Facebook. The frequent attempts to implement automated detection systems remain unsuccessful at separating hate speech from objectionable language, because user-generated content tends toward informal, brief, and diverse expressions. The determination of hate speech within texts proves exceptionally hard, since precise context detection is needed to distinguish abusive language from neutral statements. Precision in hate speech identification and filtering stands essential, because these online content forms have negative impacts on both minority and majority groups while heightening their conflicts. The research presents a stacked ensemble classification system that classifies tweets into three groups: hate speech, abusive language, or neutral. The framework uses term frequency–inverse document frequency (TF–IDF) extracted from tweet texts for which support vector machine (SVM), together with Random Forest, XGBoost, and Logistic Regression, base machine learning models function as classifiers. The final model outcome results from linking several base learning models into an ensemble configuration. The Kaggle Hate Speech data set served as training material for the system, because it contained 24,784 tweets along with eight attributes. The model performance received improvement through exclusion of manually derived features. The proposed ensemble model demonstrated superior performance with 96% accuracy, while each single classifier had lower accuracy rates (SVM: 93%, Random Forest: 94%, and XGBoost: 88%). The research outcomes show stacking represents an effective method to enhance systems for detecting hate speech operating on social media platforms.

Accurate forecasts of renewable and nonrenewable energy output are essential for meeting global energy needs and resolving environmental issues. Energy sources like the sun and wind are variable, making forecasting difficult. Changes in weather, demand, and energy policy exacerbate this unpredictability. These challenges will be addressed by the bidirectional gated recurrent unit (Bi‐GRU) model, which forecasts power‐generating outcomes more efficiently. The investigation is done over a health data set from 2000 to 2023, including the energy states of the United Kingdom, Finland, Germany, and Switzerland. The comparison of our model (Bi‐GRU) performance with other popular models, including bidirectional long short‐term memory (Bi‐LSTM), ensemble techniques combining convolutional neural networks (CNN) and Bi‐LSTM, and CNNs, make the study more interesting. The performance remains better with a mean absolute percentage error (MAPE) of 2.75%, root mean square error (RMSE) of 0.0414, mean squared error (MSE) of 0.0017, and authentify that Bi‐GRU performs much better than others. This model's superior prediction accuracy significantly enhances our ability to forecast renewable and nonrenewable energy outputs in European states, contributing to more effective energy management strategies. This study employs a bidirectional gated recurrent unit model to enhance the accuracy of energy output predictions for renewable and nonrenewable sources across the United Kingdom, Finland, Germany, and Switzerland. The findings reveal critical insights into energy production trends up to 2030, guiding strategic planning for energy management in Europe's transition toward sustainability.