Samuel Rahimeto Kebede

Electroencephalography (EEG) is a fundamental tool for monitoring brain activity and is widely used in the diagnosis and management of epilepsy. With the growing prevalence of epilepsy and the limitations of manual EEG interpretation, artificial intelligence (AI) techniques, particularly machine learning (ML) and deep learning (DL), have gained prominence in automating seizure detection, classification, and prediction. This review provides a comprehensive synthesis of AI‐based approaches applied to EEG signals for epileptic seizure analysis. We examine 86 peer‐reviewed studies from 2014 to 2023 and analyze eleven publicly available EEG datasets, highlighting their characteristics, applications, and implications for model performance. Key components of the pipeline, including feature extraction methods, model architectures (e.g., CNNs, LSTMs, and hybrids), and evaluation strategies, are critically reviewed. The paper further categorizes methodologies based on their use of spatial, temporal, and hybrid learning, as well as their capacity for patient‐specific versus patient‐independent prediction. In addition to summarizing current achievements, this survey identifies persistent challenges such as interpatient variability, lack of standardized evaluation metrics, poor model generalization across datasets, and limited real‐time deployment. It also outlines future directions, including data‐centric approaches, transfer learning, multimodal fusion, and explainable AI, all of which are crucial for translating academic research into clinically viable solutions. By critically analyzing the strengths, limitations, and opportunities in the field, this review aims to guide future research toward the development of robust, interpretable, and deployable AI systems for epileptic seizure monitoring, ultimately improving clinical decision‐making and patient outcomes.

In dynamic situations, multicriteria decision‐making (MCDM) for classification frequently takes place. The selection of suitable land sites and clear criteria is one of the most significant challenges in strategic military site selection around the world and at the national level. The main objectives of the present study are to find out the strategic military site land suitability in Adea District by using AHP, GIS, and machine learning algorithms. To determine the subject weights of criteria, a novel approach for prioritizing the intuitionistic fuzzy judgement matrix is proposed. An optimization model is developed to determine the objective weights that maximize the distance of each alternative to the negative‐ideal solution. Machine learning methods are mainly employed for filtering, interpreting, and predicting information from satellite image data. The random forest analysis method is used for patterning the military strategic site selection in the Adea District. The results of the study were able to create models and submodels of military strategic site land suitability using the weighted linear combination (WLC) mechanism. From the entire research area woreda, 87.02 km²(9.3%), 439.11 km²(46.9%), 114.59 km²(12.2%), 196.48 km²(20.9%), and 99.19 km²(10.6%) have not suitable, marginally suitable, moderately suitable, suitable, and highly suitable conditions, respectively, for a strategic military site. Changes in the six classes of land use/land cover were assessed with an overall accuracy of above 97% and an overall kappa statistic of 96%. Finally, the study actually suggests that ML, RS, GIS, and AHP algorithms are beneficial for decision‐makers (DMs) that will enhance the procedure for strategic military site, supported by multicriteria decision‐making and its applications for the study area’s land suitability. Furthermore, security and military authorities will be able to propose a planning protocol and suitable sites for strategic military site selection in the near future based on the study’s basic methodological application and main findings.

The electrical activity produced during the heartbeat is measured and recorded by an ECG. Cardiologists can interpret the ECG machine’s signals and determine the heart’s health condition and related causes of ECG signal abnormalities. However, cardiologist shortage is a challenge in both developing and developed countries. Moreover, the experience of a cardiologist matters in the accurate interpretation of the ECG signal, as the interpretation of ECG is quite tricky even for experienced doctors. Therefore, developing computer-aided ECG interpretation is required for its wide-reaching effect. 12-lead ECG generates a 1D signal with 12 channels among the well-known time-series data. Classical machine learning can develop automatic detection, but deep learning is more effective in the classification task. 1D-CNN is being widely used for CVDS detection from ECG datasets. However, adopting a deep learning model designed for computer vision can be problematic because of its massive parameters and the need for many samples to train. In many detection tasks ranging from semantic segmentation of medical images to time-series data classification, multireceptive field CNN has improved performance. Notably, the nature of the ECG dataset made performance improvement possible by using a multireceptive field CNN (MRF-CNN). Using MRF-CNN, it is possible to design a model that considers semantic context information within ECG signals with different sizes. As a result, this study has designed a multireceptive field CNN architecture for ECG classification. The proposed multireceptive field CNN architecture can improve the performance of ECG signal classification. We have achieved a 0.72 score and 0.93 AUC for 5 superclasses, a 0.46 score and 0.92 AUC for 20 subclasses, and a 0.31 score and 0.92 AUC for all the diagnostic classes of the PTB-XL dataset.

Cereals are an important and major source of the human diet. They constitute more than two-thirds of the world’s food source and cover more than 56% of the world’s cultivatable land. These important sources of food are affected by a variety of damaging diseases, causing significant loss in annual production. In this regard, detection of diseases at an early stage and quantification of the severity has acquired the urgent attention of researchers worldwide. One emerging and popular approach for this task is the utilization of machine learning techniques. In this work, we have identified the most common and damaging diseases affecting cereal crop production, and we also reviewed 45 works performed on the detection and classification of various diseases that occur on six cereal crops within the past five years. In addition, we identified and summarised numerous publicly available datasets for each cereal crop, which the lack thereof we identified as the main challenges faced for researching the application of machine learning in cereal crop detection. In this survey, we identified deep convolutional neural networks trained on hyperspectral data as the most effective approach for early detection of diseases and transfer learning as the most commonly used and yielding the best result training method.

Spider silk is an astonishingly tough biomaterial that consists almost entirely of large proteins. Studying the secrets behind the high strength nature of spider webs is very challenging due to their miniature size. In spite of their complex nature, researchers have always been inspired to mimic Nature for developing new products or enhancing the performance of existing technologies. Accordingly, the spider web can be taken as a model for optimal fiber orientation for composite materials to be used in critical structural applications. In this study an attempt is made to analyze the geometrical characteristics of the web construction building units such as spirals and radials. As a measurement tool, we have used a developed MATLAB algorithm code for measuring the node to node of rings and radials angle of orientation. Spider web image samples were collected randomly from an ecological niche with black background sample collection tools. The study shows that the radial angle of orientation is 12.7 degrees with 5 mm distance for the spirals’ mesh size. The extracted geometrical numeric values from the spider web show moderately skewed statistical data. The study sheds light on spider web utilization to develop an optimized fiber orientation reinforced composite structure for constructing, for instance, shell structures, pressure vessels and fuselage cones for the aviation industry.

A brain tumor is one of the foremost reasons for the rise in mortality among children and adults. A brain tumor is a mass of tissue that propagates out of control of the normal forces that regulate growth inside the brain. A brain tumor appears when one type of cell changes from its normal characteristics and grows and multiplies abnormally. The unusual growth of cells within the brain or inside the skull, which can be cancerous or non-cancerous has been the reason for the death of adults in developed countries and children in under developing countries like Ethiopia. The studies have shown that the region growing algorithm initializes the seed point either manually or semi-manually which as a result affects the segmentation result. However, in this paper, we proposed an enhanced region-growing algorithm for the automatic seed point initialization. The proposed approach’s performance was compared with the state-of-the-art deep learning algorithms using the common dataset, BRATS2015. In the proposed approach, we applied a thresholding technique to strip the skull from each input brain image. After the skull is stripped the brain image is divided into 8 blocks. Then, for each block, we computed the mean intensities and from which the five blocks with maximum mean intensities were selected out of the eight blocks. Next, the five maximum mean intensities were used as a seed point for the region growing algorithm separately and obtained five different regions of interest (ROIs) for each skull stripped input brain image. The five ROIs generated using the proposed approach were evaluated using dice similarity score (DSS), intersection over union (IoU), and accuracy (Acc) against the ground truth (GT), and the best region of interest is selected as a final ROI. Finally, the final ROI was compared with different state-of-the-art deep learning algorithms and region-based segmentation algorithms in terms of DSS. Our proposed approach was validated in three different experimental setups. In the first experimental setup where 15 randomly selected brain images were used for testing and achieved a DSS value of 0.89. In the second and third experimental setups, the proposed approach scored a DSS value of 0.90 and 0.80 for 12 randomly selected and 800 brain images respectively. The average DSS value for the three experimental setups was 0.86.