Abdel-Hamid Soliman

Background: Automatic detection of abnormal electroencephalogram (EEG) signals is essential for supporting clinical screening and reducing human error in EEG interpretation. Although deep learning architectures such as CNN–LSTM have shown promising performance in EEG classification, challenges related to feature variability, non-stationarity, and sensitivity to pathological patterns remain. Our previous work with windowing-based CNN-LSTM architecture achieved strong performance but it did not achieve sufficient sensitivity for reliable clinical application. Methods: To overcome these limitations, we propose an enhanced voting-based ensemble framework that combines five CNN-LSTM base classifiers with a Random Forest (RF) meta-classifier, evaluated using 10-fold cross-validation. Results: The proposed ensemble model achieved a sensitivity of 92.86%, a specificity of 72.3%, and an overall accuracy of 83%, demonstrating competitive and clinically meaningful sensitivity for abnormal EEG detection under the adopted evaluation protocol. Conclusions: These findings demonstrate that integrating multi-model feature extraction with an RF-based voting ensemble improves diagnostic reliability, reduces false negatives, and supports early and accurate detection of brain disorders. This framework not only surpasses existing approaches but also provides a flexible foundation for future advancements in clinical decision support systems.

Direction of Arrival (DOA) estimation of highly correlated or coherent sources remains a challenging problem, particularly for subspace-based methods such as MUSIC and Root-MUSIC, which require spatial smoothing or eigenvalue decomposition. This work introduces a decomposition-free DOA estimation algorithm that exploits the inherent Toeplitz structure of uniform linear arrays (ULAs). The method constructs a structured Toeplitz matrix directly from the received data vector, forms a real, symmetric compound matrix to enable real-valued processing, and extracts structured sub-blocks without matrix decomposition. A polynomial matrix is then generated, and the DOAs are obtained by solving its roots, thereby bypassing covariance decomposition entirely. This design reduces computational burden, improves numerical stability, and inherently accommodates coherent sources without the need for spatial smoothing. Computational complexity analysis shows significant savings compared to subspace methods, while MATLAB simulations across various SNR, snapshot, and array size conditions validate estimation accuracy. Furthermore, a hardware realization on the NI PXIe-7993 FPGA platform demonstrates that the decomposition-free structure requires minimal hardware resources and execution cycles, making the algorithm highly suitable for real-time embedded applications.

Breast cancer is a significant global health challenge, with highest rates of occurrence and mortality worldwide. Early detection is important for the improvement of patient outcomes and the reduction of the overall burden of the disease. Digital breast tomosynthesis (DBT) scans offer three-dimensional images of the breast tissue and is becoming a valuable tool in the detection of breast abnormalities. However, accurately classifying DBT scans is challenging due to the complexity of the anatomy of the breast and the presence of minor abnormalities. This study introduces the MSAE-DL system for the multi-class classification of DBT scans. The system incorporates a novel multi-head self-attention model with a unique ensemble classification model. Features were extracted from the Mod_AlexNet Self-Attention model and fused with histogram of oriented gradients (HOG) descriptors. Subsequently, feature vectors are reduced using three feature selection models. Finally, a novel ensemble classification model is introduced and fuses class and classifier weights for the final prediction using various classifiers. The system demonstrates optimal performance in classifying DBT scans into normal, benign, and malignant classes, achieving an accuracy of 90.13%, precision of 92.77%, and f1-score of 91.03%. The experimental results underscore the potential of this approach in enhancing DBT classification into three different classes, rather than simply binary classification.

Background: The electroencephalogram (EEG) is essential for diagnosing and classifying brain disorders, enabling early medical intervention. Its ability to identify brain abnormalities has increased its clinical use in assessing changes in brain activity. Recent advancements in deep learning have introduced effective methods for interpreting EEG signals, utilizing large datasets for enhanced accuracy. Objective: This study presents a deep learning-based model designed to classify EEG data with better accuracy compared to existing approaches. Methods: The model consists of three key components: data selection, feature extraction, and classification. Data selection employs a windowing technique, while the feature extraction and classification stages use a deep learning framework combining a convolutional neural network (CNN) and a Long Short-Term Memory (LSTM) network. The resulting architecture includes up to 18 layers. The model was evaluated using the Temple University Hospital (TUH) dataset, comprising data from 2785 patients, ensuring its applicability to real-world scenarios. Results: Comparative performance analysis shows that this approach surpasses existing methods in accuracy, sensitivity, and specificity. Conclusions: This study highlights the potential of deep learning in enhancing EEG signal interpretation, offering a pathway to more accurate and efficient diagnoses of brain disorders for clinical applications.

The emergence of Internet of Things (IoT)-based heterogeneous wireless sensor network (HWSN) technology has become widespread, playing a significant role in the development of diverse human-centric applications. The role of efficient resource utilisation, particularly energy, becomes further critical in IoT-based HWSNs than it was in WSNs. Researchers have proposed numerous approaches to either increase the provisioned resources on network devices or to achieve efficient utilisation of these resources during network operations. The application of a vast proportion of such methods is either limited to homogeneous networks or to a single parameter and limited-level heterogeneity. In this work, we propose a multi-parameter and multi-level heterogeneity model along with a cluster-head rotation method that balances energy and maximizes lifetime. This method achieves up to a 57% increase in throughput to the base station, owing to improved intra-cluster communication in the IoT-based HWSN. Furthermore, for inter-cluster communication, a mathematical framework is proposed that first assesses whether the single-hop or multi-hop inter-cluster communication is more energy efficient, and then computes the region where the next energy-efficient hop should occur. Finally, a relay-role rotation method is proposed among the potential next-hop nodes. Results confirm that the proposed methods achieve 57.44%, 51.75%, and 17.63% increase in throughput of the IoT-based HWSN as compared to RLEACH, CRPFCM, and EERPMS, respectively.

The benefits of additive manufacturing (AM) are widely recognised, boosting the AM method’s use in industry, while it is predicted AM will dominate the global manufacturing industry. Alas, 3D printing’s growth is hindered by its sustainability. AM methods generate vast amounts of residuals considered as waste, which are disposed of. Additionally, the energy consumed, the materials used, and numerous other factors render AM unsustainable. This paper aims to bring forward all documented solutions in the literature. The spotlight is on potential solutions for the Powder Bed Fusion (PBF) AM, focusing on Selective Laser Sintering (SLS), as these are candidates for mass manufacturing by industry. Solutions are evaluated critically, to identify research gaps regarding the recyclability of residual material. Only then can AM dominate the manufacturing industry, which is extremely important since this is a milestone for our transition into sustainable manufacturing. This transition itself is a complex bottleneck on our quest for becoming a sustainable civilisation. Unlike previous reviews that primarily concentrate on specific AM recycling materials, this paper explores the state of the art in AM recycling processes, incorporating the latest market data and projections. By offering a holistic and forward-looking perspective on the evolution and potential of AM, this review serves as a valuable resource for researchers and industry professionals alike.

Due to the technological advancements in wireless communication and their continuously increasing applications in collaborative and cooperative smart infrastructures, energy efficient data collection using wireless devices, has gained significant importance recently. Modern wireless sensor networks refer to network of low-powered and energy-constrained Internet of Things (IoT) devices. Although data collection using hierarchical routing with clustered network improves energy efficiency but introduces energy holes in the region closer to the data gathering center due to heavy relaying load on cluster heads. In this paper, first an improved data gathering center deployment technique for heterogeneous networks has been proposed. Technique for Order of Preference by Similarity to Ideal Solutions (TOPSIS), a multi-criteria decision-making technique, is used to determine the optimal location for the deployment of data gathering center. The proposed technique is adaptive to various shaped networks as required by IoT and increases energy efficiency. Secondly, an unequal clustering based on transmission distances has been proposed. Moreover, cubical, and spherical segmentation schemes for 3D heterogeneous networks have been proposed that assist the formation of unequal clusters. Finally, a shape independent data rate-based segmentation has been proposed that further extends the adaptability and scalability of the proposed unequal clustering. The results demonstrate that the proposed data traffic-based shape independent adaptive clustering scheme increases network lifetime up to 14.2% and 18.8% as compared to Fuzzy Logic based unequal clustering and IUCR respectively. It also reduces the overall network energy consumption up to 61.4% as compared to the state-of-the art unequal clustering methods.