Egyptian e-Learning University

Brain tumors often require treatment and multiple biopsies. They are the third most common cancer among young adults in both incidence and mortality. The expression of the O6-methylguanine-DNA methyltransferase (MGMT) gene plays an important role in predicting tumor behavior. It affects how patients respond to chemotherapy and may reduce the need for invasive procedures. Machine learning can help make accurate medical predictions, but it requires large and diverse patient datasets. These datasets are difficult to access due to privacy and legal restrictions. This article proposes a Federated Learning (FL) framework to address these challenges. FL allows different institutions to train a shared model without exchanging raw data. A hybrid deep learning model combining recurrent neural networks (RNNs) and convolutional neural networks (CNNs) is developed to analyze magnetic resonance imaging (MRI) scans from the BraTS 2021 dataset. The model aims to detect glioblastoma and predict MGMT gene expression. Two swarm intelligence algorithms, the Bayesian Search Optimization Algorithm and the Sparrow Search Optimization Algorithm, are used to optimize the model’s hyperparameters. The FL system was tested across ten universities. It performed similarly to models trained on centralized data. The proposed model, BrainGeneDeepNet, achieved high performance: 0.9758 accuracy, 0.0769 loss, 0.9980 AUC, 0.9770 recall, and 0.9782 precision. These results show that federated learning is a secure and effective approach for medical imaging and biomarker prediction.

Automatic modulation classification (AMC) is vital in cognitive radio, spectrum management, and military applications. The rapid evolution of wireless networks demands reliable and effective techniques. Traditional AMC methods, like likelihood-based and feature-based algorithms, face limits in computational complexity and generalization across different channel conditions. This paper combines Bidirectional Long Short-Term Memory (Bi-LSTM) and Convolutional Neural Networks (CNN) to form a novel hybrid deep learning model. The model exploits each branch’s strengths for temporal and spatial feature extraction. The Bi-LSTM branch identifies long-term temporal relationships, such as phase transitions, in signal sequences. In contrast, the CNN branch captures local spatial patterns like constellation structures. The proposed model outperforms baseline models—CLDNN, CNN-GRU, ResNet-LSTM, and VTCNN2—while maintaining lower computational complexity (410,762 parameters) on the RML2016.10b dataset. It achieves a classification accuracy of 92.32% at 18 dB SNR and performs robustly across low-to-high SNR regimes. These results reveal its potential for deployment in real-time wireless scenarios that are both noisy and dynamic.

Hydroxysafflor yellow A (HSYA), its primary bioactive metabolite of Carthamus tinctorius L. (safflower), has shown therapeutic potential in various inflammatory diseases. However, its role in alleviating inflammation and oxidative stress in non-alcoholic fatty liver disease (NAFLD) remains unclear. This study investigates the therapeutic effects of HSYA in mice with NAFLD, focusing on its impact on gut microbiota and serum non-targeted metabolomics to elucidate the mechanisms underlying its efficacy. NAFLD was induced in mice using a high-fat diet (HFD), followed by intragastric administration of hydroxysafflor yellow A (HSYA). Serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST), total cholesterol (TC), and triglycerides (TG) were quantified to evaluate liver function and lipid metabolism. Oxidative stress markers, including superoxide dismutase (SOD) activity and malondialdehyde (MDA) concentration, were also assessed. The pro-inflammatory cytokines IL-6, TNF-α, and IL-1β in serum were measured using ELISA. The hepatic expression of NLRP3 inflammasome and its downstream effector, Caspase-1, was analyzed by Western blot. Histopathological examination of liver tissues was performed using hematoxylin and eosin (H&E) staining to evaluate structural damage. Furthermore, alterations in the gut microbiota composition were characterized via 16S rDNA sequencing of fecal samples. Untargeted metabolomics was conducted to identify serum metabolic variations and elucidate enriched metabolic pathways associated with HSYA treatment. HSYA significantly inhibited HFD-induced weight gain and alleviated liver inflammation. It reduced serum levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and triglycerides (TG) (P < 0.05). HSYA administration decreased hepatic mRNA and protein expression of nucleotide binding oligomerization domain like receptor protein 3 (NLRP3), Caspase-1 and interleukin - 1β (IL-1β) while increasing superoxide dismutase (SOD) activity (P < 0.05). Gut microbiota analysis revealed a significant increase in the abundance of Turicibacter, while a reduction of Ruminococcus. Serum metabolomics identified a reduction in inflammation-associated metabolites, such as phenylalanine and tyrosine, alongside enhanced phenylalanine and tyrosine biosynthesis pathways. HSYA demonstrates potent anti-inflammatory and antioxidant effects, effectively mitigating liver inflammation and oxidative stress in NAFLD mice. Its therapeutic mechanisms may involve modulating gut microbiota and regulating serum phenylalanine and tyrosine metabolism, offering insights into its potential as a treatment for NAFLD.

D-Psicose (DPS) serves as an optimal sucrose substitute, providing only 0.3% of sucrose’s energy content, while exhibiting anti-inflammatory properties and inhibiting lipid synthesis. However, its efficacy in managing non-alcoholic fatty liver disease (NAFLD) remains unclear. This study employed network pharmacology and molecular docking to identify potential DPS targets for NAFLD treatment. A high-fat diet was used to induce a NAFLD mouse model, with DPS administered in drinking water at 5% (high dose DPS group, DPSH group) and 2.5% (low dose DPS group, DPSL group) concentrations. After 12 weeks, blood lipid levels, liver lipid deposition, and inflammation were evaluated to assess the therapeutic effects of DPS. To explore its underlying mechanisms, colon contents 16S rRNA sequencing and serum untargeted metabolomics were performed. Results indicated that DPS significantly reduced lipid accumulation and inflammatory damage in the livers of NAFLD mice, improving both blood lipid profiles and oxidative stress. Network pharmacology analysis revealed that DPS primarily targets pathways associated with inflammation and oxidative stress, while molecular docking suggested its potential to inhibit the NF-κB pathway activation and the expression of the receptor for advanced glycation end-products (RAGE), findings corroborated by Western blotting. Additionally, gut microbiota and serum metabolomics analyses demonstrated that DPS improved microbiota composition by increasing the abundance of beneficial bacteria, such as Akkermansia, and restored serum metabolomic balance, enhancing anti-inflammatory and antioxidant metabolites like Tretinoin and Pyridoxamine. The non-targeted metabolomics results suggest that DPS is mediated by glutathione metabolism, arginine and proline metabolism, unsaturated fatty acid biosynthesis, and linoleic acid metabolism interferes with NAFLD progression. In conclusion, DPS may alleviate oxidative stress and lipid accumulation in NAFLD mice through the AGEs/RAGE/NF-κB pathway, while also ameliorating gut microbiota dysbiosis and serum metabolomic disturbances, fostering the production of anti-inflammatory and antioxidant metabolites.

Purpose The study aims to better understand the impact of susceptibility to social influence (normative and informational) on perceived risk and the consequent impacts on attitudes towards counterfeiting and intention to purchase counterfeit brands. Design/methodology/approach A single cross-sectional descriptive research was employed, and questionnaires were used to collect data from 361 counterfeit buyers. Structural equation modelling (SEM) based on partial least squares (PLS-SEM) was applied to analyse data and test the research hypotheses. Findings Results showed that normative susceptibility to social influence significantly increased attitudes towards counterfeiting but not purchase intention; its impact on intention was mediated by perceived risk and attitudes. Although information susceptibility to social influence increased purchase intention, it had no significant impact on attitudes and perceived risk. Originality/value The current study empirically explores the relationship between susceptibility to social influence and perceived risk in the context of non-deceptive counterfeit consumption, by integrating the foundations of the theory of planned behaviour (TPB) and social cognitive theory (SCT).

The emergence of artificial intelligence (AI) and its related capabilities has led industries to rethink the existing practices of conventional supply chain management and data analysis. Machine learning (ML), Deep Learning (DL) and their unique ability to predict future data and classify data have led to important research in the supply chain (SC) domain, particularly in identifying and prioritizing supply chain risks. This paper proposes several DL methodologies to exploit the benefit of DL, particularly to identify whether any product will be delivered late due to any unforeseen reason in a complex SC system. Four different DL architectures (Simple-LSTM, Deep-LSTM, 1D-CNN, and TCN-1DSPCNN models) are proposed to extract features, while six variant classifiers: Softmax, random trees (RT), random forest (RF), K-nearest neighbor (KNN), artificial neural network (ANN), and support vector machine (SVM), were used to classify delay or non-delay information. By seamlessly capturing intricate temporal dependencies, these DL models enhance accuracy in robustly identifying supply chain late orders. Leveraging their hierarchical feature learning, these proposed DL models excel in recognizing subtle patterns and correlations, making them ideal for classifying late orders within the supply chain. Their parallel processing prowess facilitates real-time decision support, allowing organizations to address potential delays and allocate resources effectively and proactively. Five-fold cross-validation is presented to avoid over-fitting and to prove the efficiency of the proposed DL models. The total accuracies of the six ML classifiers are 74.03, 75.81, 93.35, 87.72, 93.59, and 95.10, respectively, while the maximum accuracies obtained from four proposed DL methodologies obtained an accuracy of 97.6, 98.63, 100, 100% respectively using the SVM classifier for predicting late orders based on five-fold cross-validation.

All of us know that cryptography is an innovative security strategy. Network security construction with authentication techniques of its layers through blockchain technologies is an important field to discuss. The problem that we are trying to solve is the Transaction Privacy Leakage in public Blockchain networks with IOT networks, which has resulted in the publicity of this data on the network as well as synchronizing the information that allowed it to be accessed and propagated between distributed nodes. At the same time, there are some privacy risk concerns associated with public data wherein transactions contain sensitive information about their issuers. Although some previous research introduced models to deal with the problem of Transaction Privacy Leakage like deterministic key generation, mixing services, ring signature, zero-knowledge proof, and quantum-resistant algorithms, the suggested models do not fully achieve prevention or integrity in all cases. This paper presents a developed Multi-Layer Blockchain Security Model (MLBSM) that can be used to protect IoT networks while also facilitating their implementation for protecting IoT networks and similar networks to prevent Transaction Privacy Leakage for all users in the public blockchain network. The clustering concept is utilized to facilitate the multi-layer architecture. By implementing this, we may achieve unprecedented levels of security and transparency in the blockchain network which will protect the privacy of all users in different technologies.

Detecting hate speech automatically in social media forensics has emerged as a highly challenging task due to the complex nature of language used in such platforms. Currently, several methods exist for classifying hate speech, but they still suffer from ambiguity when differentiating between hateful and offensive content and they also lack accuracy. The work suggested in this paper uses a combination of the Whale Optimization Algorithm (WOA) and Particle Swarm Optimization (PSO) to adjust the weights of two Multi-Layer Perceptron (MLPs) for neutrosophic sets classification. During the training process of the MLP, the WOA is employed to explore and determine the optimal set of weights. The PSO algorithm adjusts the weights to optimize the performance of the MLP as fine-tuning. Additionally, in this approach, two separate MLP models are employed. One MLP is dedicated to predicting degrees of truth membership, while the other MLP focuses on predicting degrees of false membership. The difference between these memberships quantifies uncertainty, indicating the degree of indeterminacy in predictions. The experimental results indicate the superior performance of our model compared to previous work when evaluated on the Davidson dataset.