Search

Dystrophinopathies are commonly affecting inherited muscular disease over the globe. Magnetic resonance imaging (MRI) is widely employed as a significant tool to diagnose dystrophinopathies. Though MRI is effective, it is mainly based on personal experiences and can simply result in misdiagnosis. This study designs a multi-objective quantum tunicate swarm optimization with deep learning (MOQTSO-DL) model to diagnose dystrophinopathies using muscle MRI images. The proposed model involves a RoI detection process by an optimized region growing approach where the initial seed points and thresholds are effectively determined by the MOQTSO algorithm. Besides, capsule network (CapsNet) is employed as a feature extractor to derive an optimal set of features. Moreover, MOQTSO with extreme learning machine (ELM) based classifier is used to allocate appropriate class labels for the muscle MRI images. The design of the MOQTSO algorithm for the initial seed point selection of RoI detection and parameter tuning of the ELM model depicts the novelty of the work. Extensive experimental analysis is carried out to showcase the improved performance of the proposed method. The simulation outcomes reported the better classification outcomes of the MOQTSO-DL method over the other compared methods. [ABSTRACT FROM AUTHOR]

A cyber physical system (CPS) is a network of cyber (computation, communication) and physical (sensors, actuators) components which interact with one another in a feedback form with human intervention. CPS authorizes the critical infrastructure and is treated as essential in day to day life as they exist in the basis of future smart devices. The increased exploitation of the CPS results in various threats and becomes a global problem. Therefore, it becomes essential to develop a safe, efficient, and robust CPS for real tine environment. For resolving this problem and accomplish security in CPS environment, intrusion detection system (IDS) can be developed. This study introduces an imbalanced generative adversarial network (IGAN) with optimal kernel extreme learning machine (OKELM), called IGAN-OKELM technique for intrusion detection in CPS environment. The proposed IGAN-OKELM technique mainly aims to address the class imbalance problem and intrusion detection. Besides, the IGAN-OKELM technique involves the IGAN model handling the class imbalance problem by the use of imbalanced data filter and convolution layers to the conventional generative adversarial network (GAN), which generates new instances for minority class labels. Moreover, the OKELM model is applied as a classifier and the optimal parameter tuning of the KELM model is performed by the use of sand piper optimization (SPO) algorithm and thereby improvises the intrusion detection performance. A wide ranging simulation analysis is carried out using benchmark dataset and the results are examined under varying aspects. The experimental results reported the better performance of the IGAN-OKELM technique over the recent state of art approaches interms of different measures. [ABSTRACT FROM AUTHOR]

Due to the advancements of high-speed networks, mobile edge computing (MEC) has received significant attention to bring processing and storage resources in client's proximity. The MEC is also a form of Edge Network or In-network computing where the resources are brought closer to the user end (edge) of the network while increasing QoE. On the other hand, the increase in the utilization of the internet of things (IoT) gadgets results in the generation of cybersecurity issues. In recent times, the advent of machine learning (ML) and deep learning (DL) techniques paves way in the detection of existing traffic conditions, data offloading, and cyberattacks in MEC. With this motivation, this study designs an effective deep learning based data offloading and cyberattack detection (DL-DOCAD) technique for MEC. The goal of the DL-DOCAD technique is to enhance the QoE in MEC systems. The proposed DL-DOCAD technique comprises traffic prediction, data offloading, and attack detection. The DL-DOCAD model applies a gated recurrent unit (GRU) based predictive model for traffic detection. In addition, an adaptive sampling cross entropy (ASCE) approach is employed for the maximization of throughput and decision making for offloading users. Moreover, the birds swarm algorithm based feed forward neural network (BSA-FFNN) model is used as a detector for cyberattacks in MEC. The utilization of BSA to appropriately tune the parameters of the FFNN helps to boost the classification performance to a maximum extent. A comprehensive set of simulations are performed and the resultant experimental values highlight the improved performance of the DL-DOCAD technique with the maximum detection accuracy of 0.992. [ABSTRACT FROM AUTHOR]

In recent days, Cognitive Cyber-Physical System (CCPS) has gained significant interest among interdisciplinary researchers which integrates machine learning (ML) and artificial intelligence (AI) techniques. This era is witnessing a rapid transformation in digital technology and AI where brain-inspired computing-based solutions will play a vital role in industrial informatics. The application of CCPS with brain-inspired computing in Industry 4.0 will create a significant impact on industrial evolution. Though the CCPSs in industrial environment offer several merits, security remains a challenging design issue. The rise of artificial intelligence AI techniques helps to address cybersecurity issues related to CCPS in industry 4.0 environment. With this motivation, this paper presents a new AI-enabled multimodal fusion-based intrusion detection system (AIMMF-IDS) for CCPS in industry 4.0 environment. The proposed model initially performs the data pre-processing technique in two ways namely data conversion and data normalization. In addition, improved fish swarm optimization based feature selection (IFSO-FS) technique is used for the appropriate selection of features. The IFSO technique is derived by the use of Levy Flight (LF) concept into the searching mechanism of the conventional FSO algorithm to avoid the local optima problem. Since the single modality is not adequate to accomplish enhanced detection performance, in this paper, a weighted voting based ensemble model is employed for the multimodal fusion process using recurrent neural network (RNN), bi-directional long short term memory (Bi-LSTM), and deep belief network (DBN), depicts the novelty of the work. The simulation analysis of the presented model highlighted the improved performance over the recent state of art techniques interms of different measures. [ABSTRACT FROM AUTHOR]

Accurate classification of brain tumor subtypes is important for prognosis and treatment. Researchers are developing tools based on static and dynamic feature extraction and applying machine learning and deep learning. However, static feature requires further analysis to compute the relevance, strength, and types of association. Recently Bayesian inference approach gains attraction for deeper analysis of static (hand-crafted) features to unfold hidden dynamics and relationships among features. We computed the gray level co-occurrence (GLCM) features from brain tumor meningioma and pituitary MRIs and then ranked based on entropy methods. The highly ranked Energy feature was chosen as our target variable for further empirical analysis of dynamic profiling and optimization to unfold the nonlinear intrinsic dynamics of GLCM features extracted from brain MRIs. The proposed method further unfolds the dynamics and to detailed analysis of computed features based on GLCM features for better understanding of the hidden dynamics for proper diagnosis and prognosis of tumor types leading to brain stroke. [ABSTRACT FROM AUTHOR]

Recently, autonomous driving systems have become hot research which allows the drivers in making decisions to enhance safety, decrease traffic accidents, and move nearer toward completely autonomous cars and intelligent transportation systems. Autonomous driving systems necessitate consistent and accurate detection technique to detect objects in the real drivable environment. Though several object detection approaches have been available in the literature, a robust technique is needed for the recognition of occluded or truncated objects. Therefore, computer vision-based approaches can be used to accomplish cost-effective and robust solutions for the object detection process. In this aspect, this study focuses on the design of robust deep learning (DL)-enabled object detection and classification (RDL-ODC) model for autonomous driving systems. Primarily, preprocessing is performed to divide the images into local patches and transform them into a compatible form. In addition, the Adam optimizer-based MobileNetv2 model is applied as a feature extractor, and linear discriminant analysis (LDA) is used to reduce the dimensionality of the features. Moreover, the optimal kernel extreme learning machine (OKELM) model is employed as a classifier. To properly tune the parameters included in the KELM method, the cuckoo search optimization (CSO) algorithm is utilized, and consequently, the overall classification accuracy gets improvised, showing the novelty of the work. A wide variety of simulation takes place on benchmark dataset, and the results are investigated in terms of different evaluation metrics. The simulation result demonstrates the promising performances of the RDL-ODC technique over the advanced methods with the maximum average precision of 0.960 and minimum average miss rate of 0.192%. [ABSTRACT FROM AUTHOR]

Accurate multi-time scale prediction of groundwater level (GWL) is important for water resources planning and management. But it is difficult to achieve reliable and robust GWL predictive even by the use of soft-computing techniques which considers the uncontrollable error, indefinite input, and unneglectable uncertainty at the time of modeling process. Soft computing approaches integrated into the data pre-treatment approach, input selection model, or uncertainty analysis can be employed to resolve this issue. The design of extensive deterministic and uncertainty analysis of automated models for the prediction of GWL is yet to be extensively explored. In this aspect, this study focuses on the design of type-II fuzzy logic system with uncertainty handling (T2FLS-UH) for GWL prediction. The goal of the T2FLS-UH technique is for predicting the GWL with the consideration of uncertainty. The T2FLS-UH technique encompasses different stages of operations such as data preparation, feature selection, prediction, and membership function selection. Besides, an ensemble empirical mode decomposition (EEMD) technique is involved for the decomposition of the original signals as to various intrinsic mode functions (IMFs). In addition, the Boruta approach is used for the selection of appropriate input variables for the prediction process. Moreover, the T2FLS model is applied for the GWL prediction process to estimate the value for every IMF. To improvise the predictive performance, the membership function of the T2FLS technique is chosen optimally by seagull optimization (SGO) algorithm. A wide range of simulations was carried to highlight the enhanced predictive performance of the T2FLS-UH technique. The experimental values pointed out the supremacy of the T2FLS-UH technique over the recent state of art prediction techniques. [ABSTRACT FROM AUTHOR]

Due to the rapid increase in exchange of text information via internet network, the security and the reliability of the digital content has become a major research issue. The main challenges faced by researchers are content authentication, integrity verification and tampering detection of digital contents. In this paper, these issues are addressed with great emphasis on text information, which is natural language dependent. Hence, a novel intelligent zero-watermarking approach is proposed for content authentication and tampering detection of Arabic text contents. In the proposed approach, both the embedding and extracting of the watermark are logically implemented, which causes no change on the digital text. This is achieved by using fourth-level-order and alphanumeric mechanism of Markov model as a soft computing technique for the analysis of Arabic text to obtain the features of the given text which is considered as the digital watermark. This digital watermark is later used for the detection of any tampering attack on the received Arabic text. An extensive set of experiments using four datasets of varying lengths proves the effectiveness of the proposed approach in terms of robustness, effectiveness and applicability under multiple random locations of insertion, reorder and deletion attacks. Compared with baseline approaches, the proposed approach has improved performance regarding watermark robustness and tampering detection accuracy. [ABSTRACT FROM AUTHOR]