Search

Background: Skin cancer is one of the highly occurring diseases in human life. Early detection and treatment are the prime and necessary points to reduce the malignancy of infections. Deep learning techniques are supplementary tools to assist clinical experts in detecting and localizing skin lesions. Vision transformers (ViT) based on image segmentation classification using multiple classes provide fairly accurate detection and are gaining more popularity due to legitimate multiclass prediction capabilities., Materials and Methods: In this research, we propose a new ViT Gradient-Weighted Class Activation Mapping (GradCAM) based architecture named ViT-GradCAM for detecting and classifying skin lesions by spreading ratio on the lesion's surface area. The proposed system is trained and validated using a HAM 10000 dataset by studying seven skin lesions. The database comprises 10 015 dermatoscopic images of varied sizes. The data preprocessing and data augmentation techniques are applied to overcome the class imbalance issues and improve the model's performance., Result: The proposed algorithm is based on ViT models that classify the dermatoscopic images into seven classes with an accuracy of 97.28%, precision of 98.51, recall of 95.2%, and an F1 score of 94.6, respectively. The proposed ViT-GradCAM obtains better and more accurate detection and classification than other state-of-the-art deep learning-based skin lesion detection models. The architecture of ViT-GradCAM is extensively visualized to highlight the actual pixels in essential regions associated with skin-specific pathologies., Conclusion: This research proposes an alternate solution to overcome the challenges of detecting and classifying skin lesions using ViTs and GradCAM, which play a significant role in detecting and classifying skin lesions accurately rather than relying solely on deep learning models., (© 2024 The Author(s). Skin Research and Technology published by John Wiley & Sons Ltd.)

Artificial Intelligence (AI) is being increasingly used for diagnosing Vision-Threatening Diabetic Retinopathy (VTDR), which is a leading cause of visual impairment and blindness worldwide. However, previous automated VTDR detection methods have mainly relied on manual feature extraction and classification, leading to errors. This paper proposes a novel VTDR detection and classification model that combines different models through majority voting. Our proposed methodology involves preprocessing, data augmentation, feature extraction, and classification stages. We use a hybrid convolutional neural network-singular value decomposition (CNN-SVD) model for feature extraction and selection and an improved SVM-RBF with a Decision Tree (DT) and K-Nearest Neighbor (KNN) for classification. We tested ourmodel on the IDRiD dataset and achieved an accuracy of 98.06%, a sensitivity of 83.67%, and a specificity of 100% for DR detection and evaluation tests, respectively. Our proposed approach outperforms baseline techniques and provides a more robust and accurate method for VTDR detection. [ABSTRACT FROM AUTHOR]

According to the World Health Organization, heart disease is the biggest cause of death worldwide. It may be possible to bring down the overall death rate of individuals if cardiovascular disease can be detected in its earlier stages. If the cardiac disease is detected at an earlier stage, there is a greater possibility that it may be successfully treated and managed under the guidance of a physician. Recent advances in areas such as the Internet of Things, cloud storage, and machine learning have given rise to renewed optimism over the capacity of technology to bring about a paradigm change on a global scale. At the bedside, the use of sensors to capture vital signs has grown increasingly commonplace in recent years. Patients are manually monitored using a monitor located at the patient's bedside; there is no automatic data processing taking place. These results, which came from an investigation of cardiovascular disease carried out across a large number of hospitals, have been used in the development of a protocol for the early, automated, and intelligent identification of heart disorders. The PASCAL data set is prepared by collecting data from different hospitals using the digital stethoscope. This data set is publicly available, and it is used by many researchers around the world in experimental work. The proposed strategy for doing research includes three steps. The first stage is known as the data collection phase, the data is collected using biosensors and IoT devices through wireless sensor networks. In the second step, all of the information pertaining to healthcare is uploaded to the cloud so that it may be analyzed. The last step in the process is training the model using data taken from already-existing medical records. Deep learning strategies are used in order to classify the sound that is produced by the heart. The deep CNN algorithm is used for sound feature extraction and classification. The PASCAL data set is essential to the functioning of the experimental environment. The deep CNN model is performing most accurately. [ABSTRACT FROM AUTHOR]

Deep neural networks (DNNs) can improve the image analysis and interpretation of remote sensing technology by extracting valuable information from images, and has extensive applications such as military affairs, agriculture, environment, transportation, and urban division. The DNNs for object detection can identify and analyze objects in remote sensing images through fruitful features of images, which improves the efficiency of image processing and enables the recognition of large-scale remote sensing images. However, many studies have shown that deep neural networks are vulnerable to adversarial attack. After adding small perturbations, the generated adversarial examples will cause deep neural network to output undesired results, which will threaten the normal recognition and detection of remote sensing systems. According to the application scenarios, attacks can be divided into the digital domain and the physical domain, the digital domain attack is directly modified on the original image, which is mainly used to simulate the attack effect, while the physical domain attack adds perturbation to the actual objects and captures them with device, which is closer to the real situation. Attacks in the physical domain are more threatening, however, existing attack methods generally generate the patch with bright style and a large attack range, which is easy to be observed by human vision. Our goal is to generate a natural patch with a small perturbation area, which can help some remote sensing images used in the military to avoid detection by object detectors and im-perceptible to human eyes. To address the above issues, we generate a rust-style adversarial patch generation framework based on style transfer. The framework takes a heat map-based interpretability method to obtain key areas of target recognition and generate irregular-shaped natural-looking patches to reduce the disturbance area and alleviates suspicion from humans. To make the generated adversarial examples have a higher attack success rate in the physical domain, we further improve the robustness of the adversarial patch through data augmentation methods such as rotation, scaling, and brightness, and finally, make it impossible for the object detector to detect the camouflage patch. We have attacked the YOLOV3 detection network on multiple datasets. The experimental results show that our model has achieved a success rate of 95.7% in the digital domain. We also conduct physical attacks in indoor and outdoor environments and achieve an attack success rate of 70.6% and 65.3%, respectively. The structural similarity index metric shows that the adversarial patches generated are more natural than existing methods. [ABSTRACT FROM AUTHOR]

News media agencies are known to publish misinformation, disinformation, and propaganda for the sake of money, higher news propagation, political influence, or other unfair reasons. The exponential increase in the use of social media has also contributed to the frequent spread of fake news. This study extends the concept of symmetry into deep learning approaches for advanced natural language processing, thereby improving the identification of fake news and propaganda. A hybrid HyproBert model for automatic fake news detection is proposed in this paper. To begin, the proposed HyproBert model uses DistilBERT for tokenization and word embeddings. The embeddings are provided as input to the convolution layer to highlight and extract the spatial features. Subsequently, the output is provided to BiGRU to extract the contextual features. The CapsNet, along with the self-attention layer, proceeds to the output of BiGRU to model the hierarchy relationship among the spatial features. Finally, a dense layer is implemented to combine all the features for classification. The proposed HyproBert model is evaluated using two fake news datasets (ISOT and FA-KES). As a result, HyproBert achieved a higher performance compared to other baseline and state-of-the-art models. [ABSTRACT FROM AUTHOR]

Artificial intelligence plays an essential role in diagnosing lung cancer. Lung cancer is notoriously difficult to diagnose until it has progressed to a late stage, making it a leading cause of cancer-related mortality. Lung cancer is fatal if not treated early, making this a significant issue. Initial diagnosis of malignant nodules is often made using chest radiography (X-ray) and computed tomography (CT) scans; nevertheless, the possibility of benign nodules leads to wrong choices. In their first phases, benign and malignant nodules seem very similar. Additionally, radiologists have a hard time viewing and categorizing lung abnormalities. Lung cancer screenings performed by radiologists are often performed with the use of computer-aided diagnostic technologies. Computer scientists have presented many methods for identifying lung cancer in recent years. Low-quality images compromise the segmentation process, rendering traditional lung cancer prediction algorithms inaccurate. This article suggests a highly effective strategy for identifying and categorizing lung cancer. Noise in the pictures was reduced using a weighted filter, and the improved Gray Wolf Optimization method was performed before segmentation with watershed modification and dilation operations. We used InceptionNet-V3 to classify lung cancer into three groups, and it performed well compared to prior studies: 98.96% accuracy, 94.74% specificity, as well as 100% sensitivity. [ABSTRACT FROM AUTHOR]

Deep Residual Networks have recently been shown to significantly improve the performance of neural networks trained on ImageNet, with results beating all previous methods on this dataset by large margins in the image classification task. However, the meaning of these impressive numbers and their implications for future research are not fully understood yet. In this survey, we will try to explain what Deep Residual Networks are, how they achieve their excellent results, and why their successful implementation in practice represents a significant advance over existing techniques. We also discuss some open questions related to residual learning as well as possible applications of Deep Residual Networks beyond ImageNet. Finally, we discuss some issues that still need to be resolved before deep residual learning can be applied on more complex problems. [ABSTRACT FROM AUTHOR]

Speaker recognition is a task that identifies the speaker from multiple audios. Recently, advances in deep learning have considerably boosted the development of speech signal processing techniques. Speaker or speech recognition has been widely adopted in such applications as smart locks, smart vehicle-mounted systems, and financial services. However, deep neural network-based speaker recognition systems (SRSs) are susceptible to adversarial attacks, which fool the system to make wrong decisions by small perturbations, and this has drawn the attention of researchers to the security of SRSs. Unfortunately, there is no systematic review work in this domain. In this work, we conduct a comprehensive survey to fill this gap, which includes the development of SRSs, adversarial attacks and defenses against SRSs. Specifically, we first introduce the mainstream frameworks of SRSs and some commonly used datasets. Then, from the perspectives of adversarial example generation and evaluation, we introduce different attack tasks, the prior knowledge of attacks, perturbation objects, perturbation constraints, and attack effect evaluation indicators. Next, we focus on some effective defense strategies, including adversarial training, attack detection, and input refactoring against existing attacks, and analyze their strengths and weaknesses in terms of fidelity and robustness. Finally, we discuss the challenges posed by audio adversarial examples in SRSs and some valuable research topics in the future. [ABSTRACT FROM AUTHOR]

Healthcare is one of the notable areas where the integration of the Internet of Things (IoT) is highly adopted, also known as the Medical IoT (MIoT). So far, MIoT is revolutionizing healthcare because it provides many advantages for the benefit of patients and healthcare personnel. The use of MIoT is becoming a booming trend, generating a large amount of IoT data, which requires proper analysis to infer meaningful information. This has led to the rise of deploying artificial intelligence (AI) technologies, such as machine learning (ML) and deep learning (DL) algorithms, to learn the meaning of this underlying medical data, where the learning process usually occurs in the cloud or telemedicine servers. Due to the exponential growth of MIoT devices and widely distributed private MIoT data sets, it is becoming a challenge to use centralized learning AI algorithms for such tasks. In this connection, federated learning (FL) is gaining traction as a possible method of learning on devices that do not need to migrate private and sensitive data to a central cloud. The terminal equipment and the central server in FL only share learning model updates to ensure that sensitive data is always kept secret. Even though this has recently become a promising research area, no other research has been conducted on this topic recently. In this paper, we synthesize recent literature and FL improvements to support FL-driven MIoT applications and services in healthcare. The findings of this research help stakeholders in academia and industry to realize the competitive advantage of the most advanced privacy preserved MIoT systems based on federal learning. [ABSTRACT FROM AUTHOR]

Alzheimer's Disease (AD) is a health apprehension of significant proportions that is negatively impacting the ageing population globally. It is characterized by neuronal loss and the formation of structures such as neurofibrillary tangles and amyloid plaques in the early as well as later stages of the disease. Neuroimaging modalities are routinely used in clinical practice to capture brain alterations associated with AD. On the other hand, deep learning methods are routinely used to recognize patterns in underlying data distributions effectively. This work uses Convolutional Neural Network (CNN) architectures in both 2D and 3D domains to classify the initial stages of AD into AD, Mild Cognitive Impairment (MCI) and Normal Control (NC) classes using the positron emission tomography neuroimaging modality deploying data augmentation in a random zoomed in/out scheme. We used novel concepts such as the blurring before subsampling principle and distant domain transfer learning to build 2D CNN architectures. We performed three binaries, that is, AD/NC, AD/MCI, MCI/NC and one multiclass classification task AD/NC/MCI. The statistical comparison revealed that 3D-CNN architecture performed the best achieving an accuracy of 89.21% on AD/NC, 71.70% on AD/MCI, 62.25% on NC/MCI and 59.73% on AD/NC/MCI classification tasks using a five-fold cross-validation hyperparameter selection approach. Data augmentation helps in achieving superior performance on the multiclass classification task. The obtained results support the application of deep learning models towards early recognition of AD. [ABSTRACT FROM AUTHOR]

Citations are references used by researchers to recognize the contributions of researchers in their articles. Citations can be used to discover hidden patterns in the research domain, and can also be used to perform various analyses in data mining. Citation analysis is a quantitative method to identify knowledge dissemination and influence papers in any research area. Citation analysis involves multiple techniques. One of the most commonly used techniques is Main Path Analysis (MPA). According to the specific use of MPA, it has evolved into various variants. Currently, MPA is carried out in different domains, but deep learning in the field of remote sensing has not yet been considered. In this paper, we have used three centrality attributes which are Degree, Betweenness and Closeness centrality to automatically identify important papers by applying clustering method based on machine learning (i.e., K-means). In addition, the main path is drawn from important papers and compared with existing manual methods. In order to conduct experiments, a data set from Web of Science (WOS) has been established, which contains 538 papers in the field of deep learning. Compared with existing works, our method provides the most relevant papers on the main path. [ABSTRACT FROM AUTHOR]

With the emergence of big data and the interest in deriving valuable insights from ever-growing and ever-changing streams of data, machine learning has appeared as an effective data analytic technique as compared to traditional methodologies. Big data has become a source of incredible business value for almost every industry. In this context, machine learning plays an indispensable role of providing smart data analysis capabilities for uncovering hidden patterns. These patterns are later translated into automating certain aspects of the decision-making processes using machine learning classifiers. This paper presents a state-of-the-art comparative analysis of machine learning and deep learning-based classifiers for multiclass prediction. The experimental setup consisted of 11 datasets derived from different domains, publicly available at the repositories of UCI and Kaggle. The classifiers include Naïve Bayes (NB), decision trees (DTs), random forest (RF), gradient boosted decision trees (GBDTs), and deep learning-based convolutional neural networks (CNN). The results prove that the ensemble-based GBDTs outperform other algorithms in terms of accuracy, precision, and recall. RF and CNN show nearly similar performance on most datasets and outperform the traditional NB and DTs. On the other hand, NB shows the lowest performance as compared to other algorithms. It is worth mentioning that DTs show the lowest precision score on the Titanic dataset. One of the main reasons is that DTs suffer from overfitting and use a greedy approach for attribute relationship analysis. [ABSTRACT FROM AUTHOR]

Crops are very important to the financial needs of a country. Due to various diseases caused by different pathogens, a large number of crops have been destroyed. As humanoids, our basic need is food for survival, and the most basic foundation of our food is agriculture. For many developing countries, it is mainly an important source of income. Bacterial diseases are one of the main diseases that cause improper production and a major economic crisis for the country. Therefore, it is necessary to detect the disease early. However, it is not easy for humans to analyze the different leaves of plants by themselves when recognizing diseases. In this article, a variety of machine learning methods are used to classify and detect leaf blight. We use the fusion of deep convolutional neural network (CNN) models obtained from Squeeze Net and Shuffle Net to improve the accuracy and robustness of large-scale applications. We use entropy to reduce the complexity of the calculation process and reduce the features in the deep learning process. In addition, we use a support vector machine (SVM) classifier to obtain the classification. We use the CIELAB color space to capture the entire color range to improve accuracy. Our results are very promising because we have achieved 98% accuracy in the early detection of leaf blight. [ABSTRACT FROM AUTHOR]

As the world keeps advancing, the need for automated interconnected devices has started to gain significance; to cater to the condition, a new concept Internet of Things (IoT) has been introduced that revolves around smart devicesʼ conception. These smart devices using IoT can communicate with each other through a network to attain particular objectives, i.e., automation and intelligent decision making. IoT has enabled the users to divide their household burden with machines as these complex machines look after the environment variables and control their behavior accordingly. As evident, these machines use sensors to collect vital information, which is then the complexity analyzed at a computational node that then smartly controls these devicesʼ operational behaviors. Deep learning-based guessing attack protection algorithms have been enhancing IoT security; however, it still has a critical challenge for the complex industries' IoT networks. One of the crucial aspects of such systems is the need to have a significant training time for processing a large dataset from the networkʼs previous flow of data. Traditional deep learning approaches include decision trees, logistic regression, and support vector machines. However, it is essential to note that this convenience comes with a price that involves security vulnerabilities as IoT networks are prone to be interfered with by hackers who can access the sensor/communication data and later utilize it for malicious purposes. This paper presents the experimental study of cryptographic algorithms to classify the types of encryption algorithms into the asymmetric and asymmetric encryption algorithm. It presents a deep analysis of AES, DES, 3DES, RSA, and Blowfish based on timing complexity, size, encryption, and decryption performances. It has been assessed in terms of the guessing attack in real-time deep learning complex IoT applications. The assessment has been done using the simulation approach and it has been tested the speed of encryption and decryption of the selected encryption algorithms. For each encryption and decryption, the tests executed the same encryption using the same plaintext for five separate times, and the average time is compared. The key size used for each encryption algorithm is the maximum bytes the cipher can allow. To the comparison, the average time required to compute the algorithm by the three devices is used. For the experimental test, a set of plaintexts is used in the simulation—password-sized text and paragraph-sized text—that achieves target fair results compared to the existing algorithms in real-time deep learning networks for IoT applications. [ABSTRACT FROM AUTHOR]

Over the last decades, load forecasting is used by power companies to balance energy demand and supply. Among the several load forecasting methods, medium-term load forecasting is necessary for grid's maintenance planning, settings of electricity prices, and harmonizing energy sharing arrangement. The forecasting of the month ahead electrical loads provides the information required for the interchange of energy among power companies. For accurate load forecasting, this paper proposes a model for medium-term load forecasting that uses hourly electrical load and temperature data to predict month ahead hourly electrical loads. For data preprocessing, modified entropy mutual information-based feature selection is used. It eliminates the redundancy and irrelevancy of features from the data. We employ the conditional restricted Boltzmann machine (CRBM) for the load forecasting. A meta-heuristic optimization algorithm Jaya is used to improve the CRBM's accuracy rate and convergence. In addition, the consumers' dynamic consumption behaviors are also investigated using a discrete-time Markov chain and an adaptive k-means is used to group their behaviors into clusters. We evaluated the proposed model using GEFCom2012 US utility dataset. Simulation results confirm that the proposed model achieves better accuracy, fast convergence, and low execution time as compared to other existing models in the literature. [ABSTRACT FROM AUTHOR]

With the rise and advancement of technology, early detection and involvement in health-associated monitoring through home control are growing with population aging. The expansion of healthy life expectations is progressively significant due to the speedy aging of the world population. The patient requires early and home-based treatment to detect and prevent disease on time and with less effort. Home-based health monitoring has been considered the need of a smart home. The services of health monitoring can facilitate the patient by collecting and analyzing the data of health for tackling diverse complex issues of health at a large scale. Health monitoring is a sustainable progression of clinical trials for ensuring that health is monitored according to the defined protocol and standard operating procedures. Various scenarios can be considered for monitoring health and are performed through experts of the field. Healthcare systems are having large-scale infrastructure of electronic devices, medical information systems, wearable and smart devices, medical records, and handheld devices. The growth in medical infrastructure, combined with the development of computational approaches in healthcare, has empowered practitioners and researchers to devise a novel solution in the innovative spectra. A detailed report of the existing literature in terms of deep learning and transfer learning is the dire need and facilitating of modern healthcare. To overcome these limitations, therefore, the proposed study presents a comprehensive review of the existing approaches, techniques, and methods associated with deep learning and transfer learning for health monitoring. This review will help researchers to formulate new ideas for facilitating healthcare based on the existing evidence. [ABSTRACT FROM AUTHOR]