Srge most important publications 2020
0 likes44 views
This report reviews the most significant publications from the Scientific Research Group in Egypt for 2019, focusing on various machine learning and optimization techniques applied to plant disease diagnosis, satellite telemetry anomaly detection, drug toxicity prediction, and agricultural crop identification. Key highlights include a convolutional neural network model optimized through particle swarm optimization, an extreme learning machine for detecting anomalies in telemetry data, and innovative models for predicting the toxicity of biotransformed drugs. The results demonstrate competitive performance across several applications, indicating the potential for advancing research and practical applications in these domains.
Srge most important publications 2020
- 1. Most Important Publications January 20 2020 This report review the most important publications published due 2019 by the Scientific Research Group in Egypt (SRGE) Publications summary
- 2. Title: An optimized model based on convolutional neural networks and orthogonal learning particle swarm optimization algorithm for plant diseases diagnosis Journal: Swarm and Evolutionary Computation (IF=6.33 & SJR=1.2 & CiteScore=7.73) Abstract: The plant disease classification based on using digital images is very challenging. In the last decade, machine learning techniques and plant images classification tools such as deep learning can be used for recognizing, detecting and diagnosing plant diseases. Currently, deep learning technology has been used for plant disease detection and classification. In this paper, an ensemble model of two pre- trained convolutional neural networks (CNNs) namely VGG16 and VGG19 have been developed for the task plant disease diagnosis by classifying the leaves images of healthy and unhealthy. In this context, CNNs are used due to its capability of overcoming the technical problems which are associated with the classification problem of plant diseases. However, CNNs suffer from a great variety of hyperparameters with specific architectures which is considered as a challenge to identify manually the optimal hyperparameters. Therefore, orthogonal learning particle swarm optimization (OLPSO) algorithm is utilized in this paper to optimize a number of these hyperparameters by finding optimal values for these hyperparameters rather than using traditional methods such as the manual trial and error method. In this paper, to prevent CNNs from falling into the local minimum and to train efficiently, an exponentially decaying learning rate (EDLR) schema is used. In this paper, the problem of the imbalanced used dataset has been solved by using random minority oversampling and random majority undersampling methods, and some restrictions in terms of both the number and diversity of samples have been overcome. The obtained results of this work show that the accuracy of the proposed model is very competitive. The experimental results are compared with the performance of other pre- trained CNN models namely InceptionV3 and Xception, whose hyperparameters were selected using a non-evolutionary method. The comparison results demonstrated that the proposed diagnostic approach has achieved higher performance than the other models. Graphic Abstract:
- 3. Title: Anomaly detection of satellite telemetry based on optimized extreme learning machine Journal: Space Safety Abstract: In aerospace, anomaly detection based on telemetry data is a critical satellite health monitoring task that is important for identifying unusual or unexpected events and for taking measurements to improve system safety and avoid serious problems. This paper introduces a novel optimized predictive model for detecting anomalies using the Grey Wolf Optimization (GWO) algorithm and an Extreme Learning Machine (ELM), called GWO-ELM. The proposed GWO-ELM is used to find anomalous events by comparing the actual observed values with the predicted intervals of telemetry data; the GWO is applied to optimize the ELM’s input weights and the bias parameters of hidden neurons to improve its prediction accuracy and ability to detect anomalies. A performance evaluation of GWO-ELM is conducted on the NASA shuttle valve benchmark dataset, which contains samples of Labeled anomalies and various metrics are collected. The experimental results for GWO-ELM show that it makes predictions with high efficiency, is stable when detecting anomalies, and requires little computational time. In addition, the results of GWO-ELM compared with those of the basic ELM algorithm with randomized parameter selection and a support vector machine (SVM), demonstrate the effectiveness and superiority of the proposed model. Graphic Abstract:
- 4. Title: Neutrosophic rule-based prediction system for toxicity effects assessment of biotransformed hepatic drugs Journal: Expert Systems with Applications (IF=4.292, SJR =1.190 & CiteScore=6.36) Abstract: Measuring toxicity is an important step in drug development. However, the current experimental methods which are used to estimate the drug toxicity are expensive and need high computational efforts. Therefore, these methods are not suitable for large-scale evaluation of drug toxicity. As a consequence, there is a high demand to implement computational models that can predict drug toxicity risks. In this paper, we used a dataset that consists of 553 drugs that biotransformed in the liver. In this data, there are four toxic effects, namely, mutagenic, tumorigenicity, irritant and reproductive effects. Each drug is represented by 31 chemical descriptors. This paper proposes two models for predicting drug toxicity risks. The proposed models consist of three phases. In the first phase, the most discriminative features are selected using rough set-based methods to reduce the classification time and improve the classification performance. In the second phase, three different sampling algorithms, namely, Random Under-Sampling, Random Over- Sampling, and Synthetic Minority Oversampling Technique (SMOTE) are used to obtain balanced data. In the third phase, the first proposed model employs the Neutrosophic Rule-based Classification System (NRCS), and the second model uses Genetic NRCS (GNRCS) to classify an unknown drug into toxic or non-toxic. The experimental results proved that the proposed models obtained high sensitivity (89–93%), specificity (91–97%), and GM (90–94%) for all toxic effects. Overall, the results of the proposed models indicate that it could be utilized for the prediction of drug toxicity in the early stages of drug development. Graphic Abstract:
- 5. Title: Toxicity risks evaluation of unknown FDA biotransformed drugs based on a multi-objective feature selection approach Journal: Applied Soft Computing (IF=4.873 & SJR=1.212 & CiteScore= 6.27) Abstract: The risk factors evaluation of the unknown biotransformed drugs is important in the drug development. However, the experimental methods that are used to perform this task are time-consuming and expensive, therefore, these methods are not suitable to assess a large dataset of drugs at the early stage of the drug development. To avoid these problems, the computational approaches can be used to predict the risk factors of the unknown biotransformed drugs. The dataset used in this study consists of 5909 drugs with 33 chemical descriptors. However, most of these descriptors are irrelevant and this may reduce the prediction accuracy; therefore, the descriptor selection approach is needed. Descriptor (Feature) selection can be considered as a multi-objective optimization problem which has two conflicting objectives, minimizing the number of the selected features and maximizing the dependency degree of the descriptors. In this paper, a new multi-objective approach is developed for the descriptor selection based on the sine- cosine algorithm and the rough set. The proposed approach consists of two stages, the feature selection stage and the predicting of an unknown drug stage. The experimental results proved that the proposed approach achieved high accuracy to all toxic effects and this indicates that it could be used for the prediction of the drug toxicity in the early stage of the drug development. Graphic Abstract:
- 6. Title: Spiking neural P grey wolf optimization system: Novel strategies for solving non-determinism problems Journal: Expert Systems with Applications (IF=4.292 & SJR=1.191 & CiteScore6.36) Abstract: Spiking neural P systems (SN P systems, in short) are the latest branch of membrane computing; inspired by the biological behavior of spiking neurons. They are considered true distributed and parallel systems; modeled to solve time consumption problem and presented the concept of parallelism usage in the computing field. This paper proposes novel strategies for solving non-determinism problem of SN P systems. The proposed algorithm relies on the parallelism feature to simulate the social hierarchy, tracking, encircling, and attacking behaviors in the grey wolf optimizer. It is modeled by collaboration between a set of SN P systems to get a feasible solution in polynomial time. Moreover, a new method named the power of signal is proposed to control the copying spikes process between neurons and differentiate between the arithmetic operations. Additionally, a time control approach is proposed to avoid non-determinism inside neurons that applied the determinism feature during firing rules. The theoretical and empirical experiments proved that the algorithm can successfully halt and in addition to the effectiveness of the proposed neural systems in getting an optimal solution in a reasonable time. As a result, this study is counted as a significant advancement in intelligent and optimization systems, whereas it has a direct impact on enhancing the performance of these systems and their applications. Graphic Abstract:
- 7. Title: Machine learning in telemetry data mining of space mission: basics, challenging and future directions Journal: Artificial Intelligence Review (IF=5.095) Abstract: The development of an intelligent artificial satellite health monitoring system is a key issue in aerospace engineering that determines satellite health status and failure using telemetry data. The modern design of data mining and machine learning technologies allows the use of satellite telemetry data and the mining of integrated information to produce an advanced health monitoring system. This paper reviews the current status and presents a framework of necessary processes on data mining to solving various problems in telemetry data such as error detection, prediction, summarization, and visualization of large quantities, and help them understand the health status of the satellite and detect the symptoms of anomalies. Machine learning technologies that include neural networks, fuzzy sets, rough sets, support vector machines, Naive Bayesian, swarm optimization, and deep learning are also presented. Also, this paper reviews a wide range of existing satellite health monitoring solutions and discusses them in the framework of remote data mining techniques. In addition, we are discussing the analysis of space debris flow analysis and the prediction of low earth orbit collision based on our orbital Petri nets model. Challenges to be addressed and future directions of research are identified and an extensive bibliography is also included. Graphic Abstract:
- 8. Title: Multi-target QSAR modelling of chemo-genomic data analysis based on Extreme Learning Machine Journal: Knowledge-Based Systems (IF=5.101 & SJR= 1.460 & CiteScore =7.01 ) Abstract: This paper presents a new Quantitative Structure-Activity Relationship (QSAR) model based on Extreme Learning Machine (ELM) to predict the biological activity of the benchmark Escape-Data sets compounds in order to provide an effective learning solution for regression analysis. The pre-processing phase of this model has been performed for the chemo-genomics datasets using the k-Nearest Neighbors (k-NN) algorithm to predict missing values of the dataset. In the second phase, the Genetic algorithm hybrid with Binary Whale Optimization algorithm (GBWOA) is adapted to determine the significance and the optimized features in feature selection phase. The min–max method is used in the third phase to transform all features to binary form in order to increases the efficiency of the proposed model by smoothing the data points and reducing fluctuation among features. ELM is used in the final phase as regression algorithm to predict chemo-genomics chemical compound. Different experiments have been performed in this paper on dataset which has been collected from ExCAPE chemo- genomics database project composed of 43509 compounds, 1134 targets besides biological activity and 40 chemical descriptors. The experimental results show that the proposed model is efficient in improving the level of prediction based on some statistical measurements. Also, ELM produced satisfactory results when the number of hidden nodes is greater than or equal to 1000. Moreover, the proposed model achieved high accuracy using R2 Graphic Abstract:
- 9. Title: A survey of swarm and evolutionary computing approaches for deep learning Journal: Artificial Intelligence Review (IF=5.095) Abstract: Deep learning (DL) has become an important machine learning approach that has been widely successful in many applications. Currently, DL is one of the best methods of extracting knowledge from large sets of raw data in a (nearly) self-organized manner. The technical design of DL depends on the feed-forward information flow principle of artificial neural networks with multiple layers of hidden neurons, which form deep neural networks (DNNs). DNNs have various architectures and parameters and are often developed for specific applications. However, the training process of DNNs can be prolonged based on the application and training set size (Gong et al. 2015). Moreover, finding the most accurate and efficient architecture of a deep learning system in a reasonable time is a potential difficulty associated with this approach. Swarm intelligence (SI) and evolutionary computing (EC) techniques represent simulation-driven non-convex optimization frameworks with few assumptions based on objective functions. These methods are flexible and have been proven effective in many applications; therefore, they can be used to improve DL by optimizing the applied learning models. This paper presents a comprehensive survey of the most recent approaches involving the hybridization of SI and EC algorithms for DL, the architecture of DNNs, and DNN training to improve the classification accuracy. The paper reviews the significant roles of SI and EC in optimizing the hyper-parameters and architectures of a DL system in context to large scale data analytics. Finally, we identify some open problems for further research, as well as potential issues related to DL that require improvements, and an extensive bibliography of the pertinent research is presented. Graphic Abstract:
- 10. Title: Analyzing Space Debris Flux and Predicting Satellites Collision Probability in LEO Orbits Based on Petri Nets Journal: IEEE ACCEESS (IF=4.098) Abstract: Space debris is rapidly becoming a real challenge and a serious threat for satellites and spacecraft motion. Most of these debris moves so fast, especially in the low earth orbits (LEOs) and so quickly that these debris can collide and penetrate the structure of the spacecraft and crash with the satellites. The National Aeronautics and Space Administration (NASA) expects that there will be an increase in the number of space particles in LEO orbits of 75% over the next 200 years if the space debris reduction measures are not followed. In this paper, we introduce the Petri net model for simulating the space debris flux estimation in the satellite orbits with respect to different debris sizes. Moreover, another Petri net model is introduced for investigating the impacts of debris flux on predicting the satellite collision probabilities. The analysis results show that there are negative correlations of the debris flux growth with the debris size d and the solar radio flux F 10.7. Moreover, the results clarify that the satellites in the LEO orbit at altitudes (h) of 600 km <; h <; 1000 km and the inclination angles (i) of 900 <; i <; 1000 cm are expected to experience more frequent collisions by 2030.
- 11. Title: Sheep Identification Using a Hybrid Deep Learning and Bayesian Optimization Approach Journal: IEEE ACCEESS (IF=4.098) Abstract: Sheep are considered a necessary source of food production worldwide. Therefore, the sheep identification is vital for managing breeding and disease. Moreover, it is the only guarantee of an individual's ownership. Therefore, in this paper, sheep identities were recognized by a deep convolutional neural network using facial bio- metrics. To obtain the best possible accuracy, different neural networks designs were surveyed and tested in this paper. The Bayesian optimization was used to automatically set the parameters for a convolutional neural network; in addition, the AlexNet configuration was also examined in this paper. In this paper, the sheep recognition algorithms were tested on a data set of 52 sheep. Not more than 10 images were taken of each sheep in different postures. Thus, the data augmentation methodologies such as rotation, reflection, scaling, blurring, and brightness modification were applied; 1000 images of each sheep were obtained for training and validation. The experiments conducted in this paper achieved an accuracy of 98%. Our approach outperforms previous approaches for sheep identification. Graphic Abstract:
- 12. Title: An improved fast fuzzy c-means using crow search optimization algorithm for crop identification in agricultural Journal: Expert Systems with Applications (IF=4.292 & SJR=1.191 & CiteScore6.36 ) Abstract: In this article is introduced an improved version of Fast Fuzzy C-Means (FFCM) by using the Crow Search optimization Algorithm (CSA) for the task of data clustering. In the proposed version of the FFCM the CSA is employed to find the centroids of the clusters that provide more accurate results in the clustering process. In this sense, the CSA avoids that the FFCM get stuck in local minima and increases the computational performance. The main feature of the CSA is its ability to find the global solution in complex optimization problems with the calibration of a reduced amount of parameter. This fact reduces the sensibility in the iterative process and permits the FFCM use the best centroids. To verify the efficiency of the proposed FFCM in a real problem, it is applied for the identification of maize plants in images from crop fields. The process starts using the adaptive colour histogram equalization to enhance the contrast and adjusts the intensities of crop images. The next task consists in to obtain the green index that is used by the CSA to compute the centroids for the FFCM algorithm. To evaluate the performance of the presented approach, it has been tested over a set of images from maize fields with different degrees of complexity, captured by different cameras and with a different perspective of the scene. Different metrics and a statistical analysis evidence that the proposed algorithm obtains better segmentation results in comparison with other methods. The experimental results proved that the proposed approach is capable of finding the optimal centroid values of the FFCM and avoids the local optima problem. Moreover, the FFCM based on CSA method can locate crop rows with the maximum accuracy as possible in digital images. Graphic Abstract:
- 13. Title: Biomechanics of artificial intervertebral disc with different materials using finite element method Journal: Soft-Computing (IF=2.367) Abstract: The main objective of this paper is to help the orthopedic in selecting the most suitable artificial intervertebral disc material that can be used in intervertebral disc replacement surgery based on the finite element method. The study uses a three-dimensional model for the lumber 4–lumber 5 spine vertebras for real patient DICOM dataset and saves as stereo lithography format. The stereo lithography file for the vertebras is converted into a computer-aided diagnostic file to be suitably imported in Solid Works software. The integrated three-dimensional model assembles and boundary conditions are applied to the model, including axial force applied on the lumber 4–lumber 5 spine. The disc is designed and constructed using Solid Works, and then four materials are applied to the artificial intervertebral disc. The results show that cobalt–chromium material is the most suitable material to be used in artificial spinal disc according to the maximum stress under an axial compression force of 300 N for cobalt–chromium and titanium was 109,308,136 N/m2 and 95,111,200 N/m2, while the maximum stress under an axial extension for cobalt–chromium and titanium was 109,308,136 N/m2 and 95,111,184 N/m2. The maximum stress under a right-side bending for cobalt– chromium and titanium was 30,456,905,728 N/m2 and 29,162,355,200 N/m2, and the maximum stress under a right-side bending for cobalt–chromium and titanium was 34,716,794,880 N/m2 and 33,086,228,480 N/m2. Graphic Abstract:
- 14. Title: Delay Tolerant Network assisted flying Ad-Hoc network scenario: modeling and analytical perspective Journal: Wireless Networks (IF=2.405) Abstract: Flying Ad-Hoc networks (FANET) are the extended paradigm of the mobile Ad-Hoc networks and, perhaps, one of the most emerging research domains in the current era. A huge number of tangible applications have been developed in this domain. The main advantages of such networks are their easy deployment, scalability, and robustness. However, the sparseness of these networks is an inherent characteristic that is known to be a bottleneck. The main objective of this work was to provide an alternative solution for the intermittently connected FANET by considering the philosophy of the Delay Tolerant Network (DTN) approach. To realize the functionality of the DTN protocols in a three- dimensional (3D) space, a social FANET model is proposed. FANET nodes are supposed to have a sparse node density. Fundamentally, the proposed DTN assisted Flying Ad hoc Network exploits the DTN routing and mobility features. The new mobility modeling for 3D spaces was re-engineered and tested with well-known routing protocols to analyze the performance of the model based on node speed, density, buffer, latency, message overhead, and power consumption. The effectiveness of 3D mobility models has also been compared against the one of classical models. The obtained results reflect a significant enhanced performance of the suggested DTN protocol for sparse FANET in a social scenario.
- 15. Title: Artificial Intelligence Technique for Gene Expression by Tumor RNA-Seq Data: A Novel Optimized Deep Learning Approach Journal: IEEE ACCEESS (IF=4.098) Abstract: Cancer is one of the most feared and aggressive diseases in the world and is responsible for more than 9 million deaths universally. Staging cancer early increases the chances of recovery. One staging technique is RNA sequence analysis. Recent advances in the efficiency and accuracy of artificial intelligence techniques and optimization algorithms have facilitated the analysis of human genomics. This paper introduces a novel optimized deep learning approach based on binary particle swarm optimization with decision tree (BPSO-DT) and convolutional neural network (CNN) to classify different types of cancer based on tumor RNA sequence (RNA-Seq) gene expression data. The cancer types that will be investigated in this research are kidney renal clear cell carcinoma (KIRC), breast invasive carcinoma (BRCA), lung squamous cell carcinoma (LUSC), lung adenocarcinoma (LUAD) and uterine corpus endometrial carcinoma (UCEC). The proposed approach consists of three phases. The first phase is preprocessing, which at first optimize the high-dimensional RNA-seq to select only optimal features using BPSO- DT and then, converts the optimized RNA-Seq to 2D images. The second phase is augmentation, which increases the original dataset of 2086 samples to be 5 times larger. The selection of the augmentations techniques was based achieving the least impact on manipulating the features of the images. This phase helps to overcome the overfitting problem and trains the model to achieve better accuracy. The third phase is deep CNN architecture. In this phase, an architecture of two main convolutional layers for featured extraction and two fully connected layers is introduced to classify the 5 different types of cancer according to the availability of images on the dataset. The results and the performance metrics such as recall, precision and F1 score show that the proposed approach achieved an overall testing accuracy of 96.90%. The comparative results are introduced, and the proposed method outperforms those in related works in terms of testing accuracy for 5 classes of cancer. Moreover, the proposed approach is less complex and consumes less memory. Graphical Abstract: