Abdelhamid Helali

This study focuses on the modeling and optimization of B_xGa_1−xAs/In_yGa_1−yAs double-junction photovoltaic cells and GaAs/B_xIn_yGa_1−x−yAs quantum well pin cells. It relies on Silvaco Atlas software 2.10.26R to simulate these structures, considering the desired number of QWs or the constant percentage of boron and indium in the III-V-based alloys of B_xGa_1−xAs, In_yGa_1−yAs, and B_xIn_yGa_1−x−yAs. The B_xIn_yGa_1−x−yAs material finds its application in the fabrication of quantum wells incorporated into the intrinsic GaAs-type absorber layer. This absorber layer is positioned between the heavily doped base and the emitting regions, in order to generate a uniform field that allows all photogenerated carriers to escape from the QWs. As part of the modeling, a comparison was made between a 40-layer GaAs/B_xIn_yGa_1−x−yAs quantum well solar cell and a standard GaAs solar cell. This analysis revealed an increase in the conversion efficiency of the simulation results from 22.66% to 31.73%. As a result, the integration of quantum wells into the structure of solar cells seems promising to optimize their performance. The optimized B_xGa_1−xAs/In_yGa_1−yAs tandem structure demonstrated a short-circuit current density of 17.50 mA/cm², an open-circuit voltage of 1.99 V, a fill factor of 87.29%, and a conversion efficiency of 30.40% under AM1.5G illumination at 300 K.

In precision oncology, the combination of the strengths of both histopathology and medical imaging provides a fertile ground for tumor characterization. Although histopathology offers a definitive cellular diagnosis, this approach is invasive and only provides a small-scale characterization of the tumor, while medical imaging modalities, such as X-ray, CT, MRI, ultrasound, and PET scans, provide a complete characterization of the tumor but, until recently, relied on the subjective ability of a human observer. The application of machine learning to radiomics aims at filling this gap, as images are mined to reveal patterns of disease not visible to the naked eye. In this perspective paper, the trajectory of machine learning in radiomics for oncology applications is critically discussed. By exploring studies using different imaging modalities, we seek to look beyond the achievements of innovative algorithms and identify the systemic weaknesses in the field, which are holding it back from translating to the clinic. In this regard, we identify two major challenges in the field: the significant effects of inter-modality and inter-scanner variability in model generalizability, and the ‘interpretability gaps’ in understanding the rationale for the decision-making process in ML algorithms. In this paper, we assert that these challenges are holding back even the best of algorithms and thus set the direction for the field in the future, advocating for the development of ML systems with emphasis on their performance in real-world settings as opposed to the lab.

The accurate detection and classification of brain tumors from magnetic resonance imaging (MRI) are critical for diagnosis and treatment planning. While deep learning has shown remarkable success in this domain, many state-of-the-art models rely on complex, end-to-end convolutional neural networks (CNNs) that require extensive computational resources and large, annotated datasets for training. This work proposes a novel and efficient methodology that, for the first time, leverages self-supervised DINO vision transformer backbones (DINO v1, DINOv2, and DINOv3) on a large corpus of natural images as powerful feature extractors for brain tumor analysis. We utilize the rich, general-purpose features from DINO-family backbones without fine-tuning the core model. These extracted features are then fed into a simpler, task-specific classifier (such as a support vector machine or a multi-layer perceptron) for the final detection and multi-class classification (e.g., glioma, meningioma, and pituitary tumor). Our methodology is evaluated on two benchmark medical imaging datasets with various classifying granularities. The results demonstrate that the proposed method achieves competitive and, in some cases, superior classification accuracy compared to representative fine-tuned convolutional neural networks and attention-based architectures, while significantly reducing the number of trainable parameters and training time. In particular, the best configuration achieves up to 98.17% accuracy and an F1-score of 98.18% on the 15-class dataset and 99.08% accuracy and an F1-score of 99.02% on the 4-class dataset. This study confirms the exceptional transfer learning capabilities of self-supervised vision transformers like DINO in the medical imaging domain, establishing it as a highly effective and efficient backbone for robust brain tumor detection and classification systems.

Computer-aided diagnosis (CAD) systems based on deep learning have shown significant potential for Alzheimer’s disease (AD) stage classification from Magnetic Resonance Imaging (MRI). Nevertheless, challenges such as class imbalance, small sample sizes, and the presence of multiple slices per subject may lead to biased evaluation and statistically unreliable performance, particularly for minority classes. In this study, a Vision Transformer (ViT)-based framework is proposed for multi-class AD classification using a Kaggle dataset containing 6400 MRI slices across four cognitive stages. A subject-wise data-splitting strategy is employed to prevent information leakage between the training and testing sets, and the statistical unreliability of near-perfect scores in underrepresented classes is critically examined. An ablation study is conducted to assess the contribution of key architectural components, demonstrating the effectiveness of self-attention and patch embedding in capturing discriminative features. Furthermore, attention-based visualization maps are incorporated to highlight brain regions influencing the model’s decisions and to illustrate subtle anatomical differences between MildDemented and VeryMildDemented cases. The proposed approach achieves a test accuracy of 97.98%, outperforming existing methods on the same dataset while providing improved interpretability. It supports early and accurate AD stage identification.

Security concerns have become paramount, especially with the increasing reliance on facial recognition technology for identity verification. The use of masks, while essential for health protection, poses a significant challenge for these systems, hindering accurate identification. This issue is particularly critical for organizations that depend on facial recognition for security and authentication, especially within IoT systems that are vulnerable to malicious activities. This article presents two major contributions to address these challenges. First, it enhances the detection of mask‐wearing individuals using a computer vision machine‐learning model, specifically a convolutional neural network (CNN). Second, it introduces a novel authentication framework for verifying users during communication between nodes and access points. The authentication system comprises three key phases. Initially, an advanced fusion technique at the access point level employs a unique hybrid biometric pattern that combines password and image features to bolster security. Next, secure transmission of this pattern between the node and the access point is ensured through AES cryptography and blockchain technology. Finally, a novel matching method compares password and image features against database records for verification, conducted during the development phase at the node level. Experimental analysis reveals that this approach delivers outstanding results, achieving 99% accuracy, 0.99 recall, 100% precision, and an F1 score of 0.989.

A brain tumor is the result of abnormal growth of cells in the central nervous system (CNS), widely considered as a complex and diverse clinical entity that is difficult to diagnose and cure. In this study, we focus on current advances in medical imaging, particularly magnetic resonance imaging (MRI), and how machine learning (ML) and deep learning (DL) algorithms might be combined with clinical assessments to improve brain tumor diagnosis. Due to its superior contrast resolution and safety compared to other imaging methods, MRI is highlighted as the preferred imaging modality for brain tumors. The challenges related to brain tumor analysis in different processes including detection, segmentation, classification, and survival prediction are addressed along with how ML/DL approaches significantly improve these steps. We systematically analyzed 107 studies (2018–2024) employing ML, DL, and hybrid models across publicly available datasets such as BraTS, TCIA, and Figshare. In the light of recent developments in brain tumor analysis, many algorithms have been proposed to accurately obtain ontological characteristics of tumors, enhancing diagnostic precision and personalized therapeutic strategies.

Summary: In the present situation, there is a huge research gap and development present in the creation and utilization of smart environment due to the development of wireless sensor network (WSN) and Internet of Things (IoT). The disadvantage of the earlier models and present systems is it consumes more power and there is a presence of network monitoring issues. Here, there is a need of developing a model which solves both this issues for the real time environment. So, in this paper, we developed a novel model which is a cross‐layered energy optimization‐based Medium Access Control (CEO‐MAC) protocol to overcome those issues. Our model considers the optimal design of the network and MAC layer. Its focus is on the CEO‐MAC routing model, cross‐layer MAC model, cross‐layer energy optimization model, and intercluster/intracluster MAC operational model. This model is briefly analyzed and simulated using Network Simulator 2 (NS2) software, which is the discrete event simulator that works using C++ and Tcl languages. Our protocol is simulated and compared with the present models such as TDL, CMISO, and PCL‐IDA protocols.