Walaa N. Ismail’s research while affiliated with Minia University and other places

Background Systemic lupus erythematosus (SLE), an autoimmune disease, predominantly affects women and is associated with an increased risk of spontaneous abortion (SA). However, traditional analytical methods found a modest relationship between some factors and SLE-SA and were limited to a small sample size, frequently associated with poor predictive performance. Objectives This study aimed to apply and evaluate an Extreme Gradient Boosting (XGBoost) model using routinely collected clinical data to identify patterns associated with spontaneous abortion in women with SLE and to identify the key variables associated with this outcome. Methods The study included adult SLE women from the Egyptian College of Rheumatology (ECR)-SLE cohort, a national multicenter study, which had available SA data. SA was defined as unexplained pregnancy loss up to 20 weeks of gestation. Patients’ demographics, clinical manifestations, SLE disease activity index (SLEDAI), therapeutic and laboratory data were used as input variables for the logistic regression (LR) and XGBoost models. We evaluated the performance of both the XGBoost and LR models by calculating the area under the receiver operating characteristic curve (AUC) for each model, and then compared these AUC values to assess which model better distinguished between patients with and without SA. The importance and direction of each variable contributing to the risk of SA were evaluated using SHapley Additive exPlanation (SHAP). Results A total of 3296 SLE women (mean ± SD age: 32.5 ± 10.1 years; median disease duration: 48 months) were included. The mean SLEDAI score was 11.3 ± 9.5. About 13.9% of the patients included had at least one abortion. Optimized XGBoost performed better (AUC 0.99) compared with LR (AUC 0.78). Positive antiphospholipid antibodies, low complement 3, longer disease duration, hypertension and the presence of mucocutaneous ulcers, as well as anticoagulants and steroid use, were among the important factors associated with SA in SLE patients. Conclusion Using information obtained in the clinical settings, the XGBoost identified variables associated with SA in women with SLE, including positive antiphospholipid antibodies, low complement 3 levels and longer disease duration. Further, longitudinal studies are necessary to evaluate the clinical utility of the proposed classification model.

Poor indoor air quality (IAQ) due to invisible pollutants in closed environments is a major threat to human health. The current monitoring systems are unable to provide precise information in real time, which is why there is a need for a smarter system of predictive IAQ management. This study presents a novel predictive system based on artificial intelligence called SLIF (Swarm-Liquid Intelligence Framework), for Smart IAQ Management. SLIF is a combination of Hybrid LSTM-GRU and Liquid Neural Networks (LiquidAdaptedNet) to predict IAQ with better accuracy, taking into consideration the dynamic environmental variations. The system integrates intelligent IAQ prediction, adaptive control, and HVAC energy-efficient optimization. It is integrated with HVAC to control air quality in real-time, and swarm intelligence methods are applied to streamline the functionality. Using evolutionary refinement techniques, such as swarm intelligence, the optimized LiquidAdaptedNet architecture achieves RMSE, MAE, R2, and MSPE values of 6.52, 6.27, 97.63%, and 1.41, respectively. New learning paradigms, including single-task, multitask, and refinement-based learning, have demonstrated significant improvements in model generalization and predictive performance. In addition to considering external parameters of the environment, the SLIF framework presents accurate forecasts for three important indoor pollutants, namely nitrogen dioxide (NO2), particulate matter (PM2.5), and carbon monoxide (CO). Depending on operating conditions, SLIF is used to achieve as much as 99.97%, 98.67%, and 94.47% of pollutant concentrations. The simulation results showed that up to 91.45% of energy savings can be achieved in the HVAC system, while maintaining good indoor air quality. Using environmental monitoring and HVAC control together with artificial intelligence, SLIF provides a scalable and dependable approach to managing the indoor air quality with the help of intelligent indoor air quality in residential, commercial, and industrial settings. This assists in making the indoor environment healthier and more sustainable.

Artificial intelligence (AI) and machine learning (ML) have made significant advances in the early detection and diagnosis of autism spectrum disorder (ASD), overcoming the limits of previous screening methods. These AI-based technologies offer more objective, scalable, and efficient methods for identifying risk behaviors associated with ASD. This article presents a novel approach for enhancing the detection and classification of ASD by integrating squeeze-and-excitation, multiscale attention mechanisms, and convolutional neural networks (CNNs) with automated hyperparameter optimization using the white shark optimization (WSO) algorithm. By leveraging attention mechanisms to focus on relevant facial features across multiple scales, this method enhances feature extraction, improves classification accuracy, and provides a robust framework for analyzing complex facial imaging data. An extensive autism dataset, encompassing both facial and multimodal datasets, was utilized in this study, including subjects from the non-ASD control (NC) group and individuals diagnosed with ASD. Experimental results demonstrate that the proposed model significantly outperforms state-of-the-art methods, achieving a high accuracy of 95.36%, precision of 92.62%, and an F1-score of 95.5% for ASD detection and classification. This proposed model is a promising tool for the accurate and early identification of ASD, which is crucial for effective treatment and management. By providing deeper insights into distinctive facial patterns and morphological features associated with ASD, the model enables physicians to make more informed decisions and develop targeted treatment plans, ultimately improving patient outcomes.

Early detection of Alzheimer’s disease (AD) is crucial, particularly in resource-constrained medical settings. This study introduces an optimized deep learning framework that conceptualizes neural networks as computational “sensors” for neurodegenerative diagnosis, incorporating feature selection, selective layer unfreezing, pruning, and algorithmic optimization. An enhanced lightweight hybrid DenseNet201 model is proposed, integrating layer pruning strategies for feature selection and bioinspired optimization techniques, including Genetic Algorithm (GA) and Harris Hawks Optimization (HHO), for hyperparameter tuning. Layer pruning helps identify and eliminate less significant features, while model parameter optimization further enhances performance by fine-tuning critical hyperparameters, improving convergence speed, and maximizing classification accuracy. GA is also used to reduce the number of selected features further. A detailed comparison of six AD classification model setups is provided to illustrate the variations and their impact on performance. Applying the lightweight hybrid DenseNet201 model for MRI-based AD classification yielded an impressive baseline F1 score of 98%. Overall feature reduction reached 51.75%, enhancing interpretability and lowering processing costs. The optimized models further demonstrated perfect generalization, achieving 100% classification accuracy. These findings underscore the potential of advanced optimization techniques in developing efficient and accurate AD diagnostic tools suitable for environments with limited computational resources.

The landscape of 5G communication introduces heightened risks from malicious attacks, posing significant threats to network security and availability. The unique characteristics of 5G networks, while enabling advanced communication, present challenges in distinguishing between legitimate and malicious traffic, making it more difficult to detect anonymous traffic. Current methodologies for intrusion detection within 5G communication exhibit limitations in accuracy, efficiency, and adaptability to evolving network conditions. In this study, we explore the application of an adaptive optimized machine learning-based framework to improve intrusion detection system (IDS) performance in wireless network access scenarios. The framework used involves developing a lightweight model based on a convolutional neural network with 11 layers, referred to as CSO-2D-CNN, which demonstrates fast learning rates and excellent generalization capabilities. Additionally, an optimized attention-based XGBoost classifier is utilized to improve model performance by combining the benefits of parallel gradient boosting and attention mechanisms. By focusing on the most relevant features, this attention mechanism makes the model suitable for complex and high-dimensional traffic patterns typical of 5G communication. As in previous approaches, it eliminates the need to manually select features such as entropy, payload size, and opcode sequences. Furthermore, the metaheuristic Cat Swarm Optimization (CSO) algorithm is employed to fine-tune the hyperparameters of both the CSO-2D-CNN and the attention-based XGBoost classifier. Extensive experiments conducted on a recent dataset of network traffic demonstrate that the system can adapt to both binary and multiclass classification tasks for high-dimensional and imbalanced data. The results show a low false-positive rate and a high level of accuracy, with a maximum of 99.97% for multilabel attack detection and 99.99% for binary task classification, validating the effectiveness of the proposed framework in the 5G wireless context.

In this research, the authors combine multiobjective evaluation metrics in the (1 + 1) evolutionary strategy with three novel methods of the Pareto optimal procedure to address the learning-to-rank (LTR) problem. From the results obtained, the Cauchy distribution as a random number generator for mutation step sizes outperformed the other distributions used. The aim of using the chosen Pareto optimal methods was to determine which method can give a better exploration–exploitation trade-off for the solution space to obtain the optimal or near-optimal solution. The best combination for that in terms of winning rate is the Cauchy distribution for mutation step sizes with method 3 of the Pareto optimal procedure. Moreover, different random number generators were evaluated and analyzed versus datasets in terms of NDCG@10 for testing data. It was found that the Levy generator is the best for both the MSLR and the MQ2007 datasets, while the Gaussian generator is the best for the MQ2008 dataset. Thus, random number generators clearly affect the performance of ES-Rank based on the dataset used. Furthermore, method 3 had the highest NDCG@10 for MQ2008 and MQ2007, while for the MSLR dataset, the highest NDCG@10 was achieved by method 2. Along with this paper, we provide a Java archive for reproducible research.

Chronic kidney disease (CKD) refers to impairment of the kidneys that may worsen over time. Early detection of CKD is crucial for saving millions of lives. As a result, several studies are currently focused on developing computer-aided systems to detect CKD in its early stages. Manual screening is time-consuming and subject to personal judgment. Therefore, methods based on machine learning (ML) and automatic feature selection are used to support graders. The goal of feature selection is to identify the most relevant and informative subset of features in a given dataset. This approach helps mitigate the curse of dimensionality, reduce dimensionality, and enhance model performance. The use of natural-inspired optimization algorithms has been widely adopted to develop appropriate representations of complex problems by conducting a blackbox optimization process without explicitly formulating mathematical formulations. Recently, snake optimization algorithms have been developed to identify optimal or near-optimal solutions to difficult problems by mimicking the behavior of snakes during hunting. The objective of this paper is to develop a novel snake-optimized framework named CKD-SO for CKD data analysis. To select and classify the most suitable medical data, five machine learning algorithms are deployed, along with the snake optimization (SO) algorithm, to create an extremely accurate prediction of kidney and liver disease. The end result is a model that can detect CKD with 99.7% accuracy. These results contribute to our understanding of the medical data preparation pipeline. Furthermore, implementing this method will enable health systems to achieve effective CKD prevention by providing early interventions that reduce the high burden of CKD-related diseases and mortality.

There are several major threats to crop production. As herbicide use has become overly reliant on weed control, herbicide-resistant weeds have evolved and pose an increasing threat to the environment, food safety, and human health. Convolutional neural networks (CNNs) have demonstrated exceptional results in the analysis of images for the identification of weeds from crop images that are captured by drones. Manually designing such neural architectures is, however, an error-prone and time-consuming process. Natural-inspired optimization algorithms have been widely used to design and optimize neural networks, since they can perform a blackbox optimization process without explicitly formulating mathematical formulations or providing gradient information to develop appropriate representations and search paradigms for solutions. Harris Hawk Optimization algorithms (HHO) have been developed in recent years to identify optimal or near-optimal solutions to difficult problems automatically, thus overcoming the limitations of human judgment. A new automated architecture based on DenseNet-121 and DenseNet-201 models is presented in this study, which is called “DenseHHO”. A novel CNN architecture design is devised to classify weed images captured by sprayer drones using the Harris Hawk Optimization algorithm (HHO) by selecting the most appropriate parameters. Based on the results of this study, the proposed method is capable of detecting weeds in unstructured field environments with an average accuracy of 98.44% using DenseNet-121 and 97.91% using DenseNet-201, the highest accuracy among optimization-based weed-detection strategies.

Alzheimer’s disease (AD) is a neurodegenerative disease that affects a large number of people across the globe. Even though AD is one of the most commonly seen brain disorders, it is difficult to detect and it requires a categorical representation of features to differentiate similar patterns. Research into more complex problems, such as AD detection, frequently employs neural networks. Those approaches are regarded as well-understood and even sufficient by researchers and scientists without formal training in artificial intelligence. Thus, it is imperative to identify a method of detection that is fully automated and user-friendly to non-AI experts. The method should find efficient values for models’ design parameters promptly to simplify the neural network design process and subsequently democratize artificial intelligence. Further, multi-modal medical image fusion has richer modal features and a superior ability to represent information. A fusion image is formed by integrating relevant and complementary information from multiple input images to facilitate more accurate diagnosis and better treatment. This study presents a MultiAz-Net as a novel optimized ensemble-based deep neural network learning model that incorporate heterogeneous information from PET and MRI images to diagnose Alzheimer’s disease. Based on features extracted from the fused data, we propose an automated procedure for predicting the onset of AD at an early stage. Three steps are involved in the proposed architecture: image fusion, feature extraction, and classification. Additionally, the Multi-Objective Grasshopper Optimization Algorithm (MOGOA) is presented as a multi-objective optimization algorithm to optimize the layers of the MultiAz-Net. The desired objective functions are imposed to achieve this, and the design parameters are searched for corresponding values. The proposed deep ensemble model has been tested to perform four Alzheimer’s disease categorization tasks, three binary categorizations, and one multi-class categorization task by utilizing the publicly available Alzheimer neuroimaging dataset. The proposed method achieved (92.3 ± 5.45)% accuracy for the multi-class-classification task, significantly better than other network models that have been reported.

Alzheimer’s disease (AD) is a neurological disease that affects numerous people. The condition causes brain atrophy, which leads to memory loss, cognitive impairment, and death. In its early stages, Alzheimer’s disease is tricky to predict. Therefore, treatment provided at an early stage of AD is more effective and causes less damage than treatment at a later stage. Although AD is a common brain condition, it is difficult to recognize, and its classification requires a discriminative feature representation to separate similar brain patterns. Multimodal neuroimage information that combines multiple medical images can classify and diagnose AD more accurately and comprehensively. Magnetic resonance imaging (MRI) has been used for decades to assist physicians in diagnosing Alzheimer’s disease. Deep models have detected AD with high accuracy in computing-assisted imaging and diagnosis by minimizing the need for hand-crafted feature extraction from MRI images. This study proposes a multimodal image fusion method to fuse MRI neuroimages with a modular set of image preprocessing procedures to automatically fuse and convert Alzheimer’s disease neuroimaging initiative (ADNI) into the BIDS standard for classifying different MRI data of Alzheimer’s subjects from normal controls. Furthermore, a 3D convolutional neural network is used to learn generic features by capturing AlD biomarkers in the fused images, resulting in richer multimodal feature information. Finally, a conventional CNN with three classifiers, including Softmax, SVM, and RF, forecasts and classifies the extracted Alzheimer’s brain multimodal traits from a normal healthy brain. The findings reveal that the proposed method can efficiently predict AD progression by combining high-dimensional MRI characteristics from different public sources with an accuracy range from 88.7% to 99% and outperforming baseline models when applied to MRI-derived voxel features.