Search

Abstract Diabetic retinopathy (DR) significantly burdens ophthalmic healthcare due to its wide prevalence and high diagnostic costs. Especially in remote areas with limited medical access, undetected DR cases are on the rise. Our study introduces an advanced deep transfer learning-based system for real-time DR detection using fundus cameras to address this. This research aims to develop an efficient and timely assistance system for DR patients, empowering them to manage their health better. The proposed system leverages fundus imaging to collect retinal images, which are then transmitted to the processing unit for effective disease severity detection and classification. Comprehensive reports guide subsequent medical actions based on the identified stage. The proposed system achieves real-time DR detection by utilizing deep transfer learning algorithms, specifically VGGNet. The system’s performance is rigorously evaluated, comparing its classification accuracy to previous research outcomes. The experimental results demonstrate the robustness of the proposed system, achieving an impressive 97.6% classification accuracy during the detection phase, surpassing the performance of existing approaches. Implementing the automated system in remote areas has transformed healthcare dynamics, enabling early, cost-effective DR diagnosis for millions. The system also streamlines patient prioritization, facilitating timely interventions for early-stage DR cases.

Abstract With easy access to social media platforms, spreading fake news has become a growing concern today. Classifying fake news is essential, as it can help prevent its negative impact on individuals and society. In this regard, an end-to-end framework for fake news detection is developed by utilizing the power of adversarial training to make the model more robust and resilient. The framework is named "ANN: Adversarial News Net," emoticons have been extracted from the datasets to understand their meanings concerning fake news. This information is then fed into the model, which helps to improve its performance in classifying fake news. The performance of the ANN framework is evaluated using four publicly available datasets, and it is found to outperform baseline methods and previous studies after adversarial training. Experiments show that Adversarial Training improved the performance by 2.1% over the Random Forest baseline and 2.4% over the BERT baseline method in terms of accuracy. The proposed framework can be used to detect fake news in real-time, thereby mitigating its harmful effects on society.

Virtual reality (VR) and immersive technology have emerged as powerful tools with numerous applications. VR technology creates a computer-generated simulation that immerses users in a virtual environment, providing a highly realistic and interactive experience. This technology finds applications in various fields, including gaming, healthcare, education, architecture, and training simulations. Understanding user immersion levels in VR is crucial and challenging for optimizing the design of VR applications. Immersion refers to the extent to which users feel absorbed and engrossed in the virtual environment. This research primarily aims to detect user immersion levels in VR using an efficient machine-learning model. We utilized a benchmark dataset based on user experiences in VR environments to conduct our experiments. Advanced deep and machine learning approaches are applied in comparison. We proposed a novel technique called Polynomial Random Forest (PRF) for feature generation mechanisms. The proposed PRF approach extracts polynomial and class prediction probability features to generate a new feature set. Extensive research experiments show that random forest outperformed state-of-the-art approaches, achieving a high immersion level detection rate of 98%, using the proposed PRF technique. We applied hyperparameter optimization and cross-validation approaches to validate the performance scores. Additionally, we utilized explainable artificial intelligence (XAI) to interpret the reasoning behind the decisions made by the proposed model for user immersion level detection in VR. Our research has the potential to revolutionize user immersion level detection in VR, enhancing the design process.

The rise in stable coronary artery disease (CAD) due to improved survival rates and population growth has increased patient numbers, straining healthcare systems. Machine learning (ML) models are being developed to predict and identify individual risk factors for early treatment, reducing harm to individuals and families. These models can predict hospitalizations, enable close monitoring of high-risk patients, and optimize medical care. Researchers are developing robust models based on ML algorithms and real-world clinical data to aid in early detection, contributing to AI research in healthcare. Advanced ML models analyze medical imaging, genetic markers, lifestyle, and environmental factors to accurately predict coronary heart disease (CHD) start and progression. Our research introduces four novel models based on two-class Logistic Regression (two-class LR), two-class Neural Network (two-class NN), two-class Decision Jungle (two-class DJ), and two-class Boosted DT (two-class BDT). Our comparative analysis reveals that the two-class Boosted DT model is the most effective, achieving an AUC score of 0.991. This model excels in real-time monitoring by predicting minor changes in patient’s health markers, allowing for timely adjustments in treatment plans. It optimizes medication selection, dosing, and intervention timing based on patient characteristics, improving therapeutic efficacy and reducing side effects. The study reveals the transformative potential of these advanced ML models in CAD prediction and management. By focusing on feature selection, algorithm improvement, and integration, our models analyze medical imaging, genetic markers, lifestyle, and environmental factors to accurately predict the onset and progression of CHD. This research proposes valuable insights into the capabilities of these models to revolutionize disease detection and management, ensuring reliable and timely healthcare interventions across various datasets.

According to the International Diabetes Federation (IDF), roughly 33% of individuals affected by diabetes exhibit diagnoses encompassing diverse severity of diabetic retinopathy. In the year 2020, approximately 463 million adults within the age bracket of 20 to 79 were documented as diabetes sufferers on a global scale. Projections suggest a rise to 700 million by 2045. The proposed automated diabetic retinopathy detection methods aim to reduce the workload of ophthalmologists. The study presents the Robust Fuzzy Local Information K-Means Clustering algorithm, an advanced iteration of the classical K-means clustering approach, integrating localized information parameters tailored to individual clusters. Comparative analysis is conducted between the performance of Robust Fuzzy Local Information K-Means Clustering and Modified Fuzzy C Means clustering, which incorporates a median adjustment parameter to augment Fuzzy C Means for diabetic retinopathy detection. The results are evaluated on three datasets: IDRiD, Kaggle, and fundus images collected from Shiva Netralaya Center, India. Achieving a 94.4% accuracy rate and an average execution time of 17.11 seconds, the proposed algorithm aims to categorize a substantial volume of retinal images, thereby improving performance and meeting the crucial demand for prompt and precise diagnoses in diabetic retinopathy healthcare.

Recent sign language skeletal-based feature models (SLSm) consist of various distracting coordinates that lead to complex deep-learning modeling. However, SLSm is not purely a spatial-temporal coordinate arrangement problem; it is also limited by human dynamics and feature aggregations. The objectives of this work are twofold: 1) to transform the skeletal features of the SLSm model to address the problem of variations in viewpoint and changes across features of repeated signs due to human dynamic; and 2) to exploit the potential of exhaustive searching in dropping distracting features to prevent complex deep learning modeling. Method: We propose a transformed skeletal feature-based model (SCT) from a feature thresholding theory. We first extract the hand-skeletal joint-related features relevant to the coordinates and positions of the hand transcription that efficiently capture human dynamics. The extracted features are transformed into a subset of a predefined threshold and fed into the proposed ensemble exhaustive feature searching. The searched features are transformed into their equivalent deep input image sequences. Outcomes: By leveraging the skeletal-based transformed and deep spatial features, the proposed method demonstrates robust performance in sign language recognition, surpassing recent deep learning models in accuracy and simplicity. The proposed skeletal features demonstrate superiority in learning complex hand gestures of public data sets, improving accuracy by more than 2%.

Alzheimer’s disease is a neurodegenerative disease causing memory loss and brain protein accumulation. Early diagnosis is crucial for clinical trials and patient care. Magnetic resonance imaging (MRI) methods have improved diagnosis and prognosis, but doctors need to interpret images proficiently. Deep learning technology has shown potential in detecting Alzheimer’s disease, but the disease progresses slower in early phases. A new dual-attention convolutional autoencoder model is presented, offering improved detection abilities and potential for real-time use in Alzheimer’s disease diagnosis. The study utilized two datasets: the first ADNI dataset, which includes three classes (MCI, CN, and AD), and the second Alzheimer’s Disease Neuroimaging Dataset, which includes two distinct classes (AD and MCI). We analyze the effectiveness of our proposed model by evaluating key performance metrics such as accuracy, precision, sensitivity, specificity, F1 score, and AUC score. In addition, we utilize cross-validation and mean absolute error to validate our model while also fine-tuning the parameters. Based on experimental data, the proposed model accurately detected Alzheimer’s disease with an accuracy of 0.9902 ± 0.0139. Based on the results, the proposed model demonstrates excellent performance compared to the existing methods described in the literature. The proposed mode achieves precision, sensitivity, and specificity of 0.9882 ± 0.0587, 0.9898 ± 0.0865, 0.9912 ± 0.0872 respectively. The model achieved an AUC score of 0.9992 for MCI and 0.9919 for AD class. Furthermore, the proposed method can enhance the affordability of Alzheimer’s disease diagnostics and increase the rate of early AD detection by facilitating remote healthcare.

There is a widespread increase of adoption of e-learning platforms, there is a great need to develop effective methods for assessing students’ cognitive abilities to deal with that environments. Traditional assessment methods need to improve in terms of analyzing the diverse range of skills and knowledge that students acquire during online learning. The current state of research work considers different levels of Bloom’s Taxonomy, which are used to evaluate students’ performance based on given text analysis. For this purpose, the Software Engineering course domain is considered. An online test was conducted among graduate university level students, consisting of 12 subjective-type questions, where 2 questions chosen to be tested at each level, concluded as 12 questions per student at accumulated levels of taxonomy. There exists Approx. 300 students’, who attempted the test and hence their textual responses are being used to evaluate the system. The methodology deploys the SVM classifier to predict the level of Bloom’s Taxonomy for exam questions and then, multi-agent system is developed to match and identify each level assigned to an agent. Each instruction based agent is trained on Random Forest Classifier to evaluate students’ cognitive skill on a scale of outcome as Good, Bad and Average. Tools, techniques, technology missing. The outcomes show that the proposed model performs much better than the existing models, provided with 98% accuracy for SVM Classifier and 92% accuracy for Random Forest Classifier to assess students’ textual responses. The novelty of this research work spins around agent prediction of cognitive level for assessment by a use of multi-agent system to recognize students’ strengths and weakness as per six levels of Bloom’s Taxonomy.

Presently, the world is progressing towards the notion of smart and secure cities. The automatic recognition of human activity is among the essential landmarks of smart city surveillance projects. Moreover, classifying group activity and behavior detection is complex and indistinct. Consequently, behavior classification systems reliant on visual data hold expansive utility across a spectrum of domains, including but not limited to video surveillance, human-computer interaction, and the safety infrastructure of smart cities. However, automatic behavior classification poses a significant challenge in the context of live videos captured by the smart city surveillance system. In this regard, the use of pictures with pre-trained convolution neural networks (CNNs)-assisted transfer learning (TL) has emerged as a potential technique for deep neural networks (DNNs) object detection., resulting in increased performance in localization for smart city surveillance. Against this backdrop, this paper explores various strategies to develop advanced synthetic datasets that could enhance accuracy when trained with modern DNNs for object detection (mAP). TL was employed to address the limitation of DL that necessitates a huge dataset. The KITTI datasets were used to train a contemporary DNN single-shot multiple box detector (SSMD) in TensorFlow. A variety of metrics were employed to assess the efficacy of the novel automated Transfer Learning (TL) system within a real-world context, specifically designed for object detection within the DL framework (referred to as OD-SSMD). The results unveiled that this developed system outperformed preceding investigations, demonstrating superior performance. Notably, it exhibited the remarkable capability to autonomously discern and pinpoint various attributes and entities within digital images, effectively identifying and localizing each item present within the images.

Digital image collections are becoming increasingly popular due to their ease of use. Still, the need for adequate indexing information makes it difficult for users to find the specific images they need. With the vast number of digital images generated daily, these databases have become enormous, making accurate image retrieval challenging. One of the most challenging tasks in computer vision and multimedia research is image annotation, where keywords are assigned to an image. Unlike humans, computers can measure colors, textures, and shapes of images but fail to interpret them semantically, known as the semantic gap. This makes image annotation complex. For semantic-level concepts generation the raw image pixels provide not enough Unmistakable information. Which mean for of “words” or “sentences” there is no clear definition with the semantics of an image unlike text annotation. Therefore, this study aims to bridge the semantic gap between low-level computer features and human interpretation of images. The proposed enhanced automatic image annotation system maps multiple labels or into single image, providing an in-depth understanding of the visual content’s meaning. This is achieved by combining Convolutional Neural Networks-based multiple features (Y is the green component of the color, Cb and Cr is the blue component and red component called YCbCr color space and Gaussian–Laplacian Pyramid) and neighbors to recall and balance precision. The image annotation (IA) scheme uses a Global Vectors for Word Representation (GloVe) model with CNN-Gaussian–Laplacian Pyramid and learning representation to predict image annotation (IA) accurately. The proposed image annotation (IA) system was execution on three public datasets and showed excellent flexibility of annotation, improved accuracy, and reduced computational costs compared to existing state-of-the-art methods. The image annotation (IA) framework can provide immense benefits in accurately selecting and extracting image features, minimizing computational complexity and facilitating annotation.

In recent years, the realms of computer vision and deep learning have ushered in transformative changes across various domains. Among these, deep learning stands out for its remarkable capacity to handle vast datasets, revolutionizing numerous fields, including the biomedical sector. In particular, its prowess has been harnessed in the realm of brain tumor identification through MRI scans, yielding impressive results. This research project is dedicated to conducting an exhaustive exploration of existing endeavors in the domain of brain tumor identification and classification via MRI scans. This endeavor is poised to be of profound value to researchers looking to leverage their deep learning expertise in the realm of brain tumor detection and categorization. The initial phase involves an overview of prior studies that have employed deep learning for categorising and detecting brain tumors. Subsequently, a meticulous analysis of deep learning studies proposed in research publications spanning (2019 to 2022) is presented in tabular form. The conclusion section comprehensively assesses the merits and demerits inherent in deep neural networks. The insights gleaned from this study promise to equip future researchers with a holistic perspective on current research trends and a nuanced understanding of the effectiveness of diverse deep learning methodologies. It is our fervent belief that this research will significantly advance the understanding of brain tumors and their detection methodologies.

Lack of an effective early sign language learning framework for a hard-of-hearing population can have traumatic consequences, causing social isolation and unfair treatment in workplaces. Alphabet and digit detection methods have been the basic framework for early sign language learning but are restricted by performance and accuracy, making it difficult to detect signs in real life. This article proposes an improved sign language detection method for early sign language learners based on the You Only Look Once version 8.0 (YOLOv8) algorithm, referred to as the intelligent sign language detection system (iSDS), which exploits the power of deep learning to detect sign language-distinct features. The iSDS method could overcome the false positive rates and improve the accuracy as well as the speed of sign language detection. The proposed iSDS framework for early sign language learners consists of three basic steps: (i) image pixel processing to extract features that are underrepresented in the frame, (ii) inter-dependence pixel-based feature extraction using YOLOv8, (iii) web-based signer independence validation. The proposed iSDS enables faster response times and reduces misinterpretation and inference delay time. The iSDS achieved state-of-the-art performance of over 97% for precision, recall, and F1-score with the best mAP of 87%. The proposed iSDS method has several potential applications, including continuous sign language detection systems and intelligent web-based sign recognition systems.

Oil spills are a paramount and immediate challenge affecting marine ecosystems globally. Effective and timely monitoring tools, such as oil detection indices, offer a swift means to track oil spill spread across vast oceanic expanses. Moreover, these indices enhance data clarity, making it more conducive for machine learning and deep learning algorithms.This study leverages the natural seepage occurring around Qaruh Island, Kuwait as a unique context for the spectral analysis of oil spills using Sentinel-2 multispectral imagery due to repeated occurrences in the same region. This research evaluated 859 single band and 455 multichannel combinations to identify the most effective combinations in oil-water separability, employing the Jeffries-Matusita (JM) distance measure as a key metric. Bands 1, 2, 3, 8A, 11, and 12 consistently featured among the top-performing indices combinations B1−B11B1+B11;B1+B2B3+B11;B1+B2B3+B12;B1+B2B3+B8A affirming the significant effect of oil spills on visible, Near-Infrared (NIR), and Shortwave Infrared (SWIR) bands. Notably, the indices developed in this study outperformed those from prior research in terms of suitability to unsupervised classification algorithms. A significant conclusion of this study is that incorporating a higher number of bands in the analysis did not correlate with an increase in JM values, suggesting that the selection of specific, informative bands is more critical than the volume of input data. These findings underscore the indispensable role of Sentinel-2 imagery in environmental investigations and highlight the potential for focused, efficient analysis using strategic band combinations for effective oil spill detection. • This study identified optimized Sentinel-2 band combinations for oil-water separability, benefiting from naturally occurring spills around Qaruh Island. • The proposed indices outperformed the previous indices for oil spill visualization and clustering. • The new indices highlighted the critical role of specific band selection over the volume of input data for effective oil spill detection.

Brain tumors present a significant medical challenge, demanding accurate and timely diagnosis for effective treatment planning. These tumors disrupt normal brain functions in various ways, giving rise to a broad spectrum of physical, cognitive, and emotional challenges. The daily increase in mortality rates attributed to brain tumors underscores the urgency of this issue. In recent years, advanced medical imaging techniques, particularly magnetic resonance imaging (MRI), have emerged as indispensable tools for diagnosing brain tumors. Brain MRI scans provide high-resolution, non-invasive visualization of brain structures, facilitating the precise detection of abnormalities such as tumors. This study aims to propose an effective neural network approach for the timely diagnosis of brain tumors. Our experiments utilized a multi-class MRI image dataset comprising 21,672 images related to glioma tumors, meningioma tumors, and pituitary tumors. We introduced a novel neural network-based feature engineering approach, combining 2D convolutional neural network (2DCNN) and VGG16. The resulting 2DCNN-VGG16 network (CVG-Net) extracted spatial features from MRI images using 2DCNN and VGG16 without human intervention. The newly created hybrid feature set is then input into machine learning models to diagnose brain tumors. We have balanced the multi-class MRI image features data using the Synthetic Minority Over-sampling Technique (SMOTE) approach. Extensive research experiments demonstrate that utilizing the proposed CVG-Net, the k-neighbors classifier outperformed state-of-the-art studies with a k-fold accuracy performance score of 0.96. We also applied hyperparameter tuning to enhance performance for multi-class brain tumor diagnosis. Our novel proposed approach has the potential to revolutionize early brain tumor diagnosis, providing medical professionals with a cost-effective and timely diagnostic mechanism.

The Gulf of Suez region accommodates diverse activities, including oil exploration and production, recreational activities, and export and import ports. The Gulf region is exposed to pollution risks due to these interactions, with few research studies documenting these pollution cases. This research aimed to use Synthetic Aperture Radar (SAR) satellite data to detect and map all the oil pollution incidents within the geographical extent of the Gulf of Suez that occurred from 2017 to 2021, locating the most affected regions and possible sources of pollution. It enabled the detection and mapping of nearly 150 oil spill incidents that occurred over 67 dates during the study period and covered 851 km2 of the sea surface. The year 2018 recorded the greatest pollution area over the study period, with 201 km2. Along the Gulf coast, Suez, Ain Sokhna, and Ras Ghareb cities recorded the highest number of marine pollution incidents. The research also located seven sources of pollution that frequently discharge into the Gulf water without regulations. This research recommends implementing a real-time monitoring system for oil pollution to robustly detect any future oil incidents in these high-risk areas as quickly as possible and minimize their environmental impact.

Smart technologies are advancing the development of cutting-edge systems by exploring the future network. The Internet of Things (IoT) and many multimedia sensors interact with each other for collecting and transmitting visual data. However, managing enormous amounts of data from numerous network devices is one of the main research challenges. In this context, various IoT systems have been investigated and have provided efficient data retrieval and processing solutions. For multimedia systems, however, controlling inefficient bandwidth utilization and ensuring timely transmission of vital information are key research concerns. Moreover, to transfer multimedia traffic while balancing communication costs for the IoT system, a sustainable solution with intelligence in real-life applications is demanded. Furthermore, trust must be formed for technological advancement to occur; such an approach provides the smart communication paradigm with the incorporation of edge computing. This study proposed a model for optimizing multimedia using a combination of edge computing intelligence and authentic strategies. Mobile edges analyze network states to discover the system's status and minimize communication disruptions. Moreover, direct and indirect authentication determines the reliability of data forwarders and network stability. The proposed authentication approach minimizes the possibility of data compromise and increases trust in multimedia surveillance systems. Using simulation testing, the proposed model outperformed other comparable work in terms of byte delivery, packet overhead, packet delay, and data loss metrics.

Abstract This paper explores the concept of smart cities and the role of the Internet of Things (IoT) and machine learning (ML) in realizing a data-centric smart environment. Smart cities leverage technology and data to improve the quality of life for citizens and enhance the efficiency of urban services. IoT and machine learning have emerged as key technologies for enabling smart city solutions that rely on large-scale data collection, analysis, and decision-making. This paper presents an overview of smart cities’ various applications and discusses the challenges associated with implementing IoT and machine learning in urban environments. The paper also compares different case studies of successful smart city implementations utilizing IoT and machine learning technologies. The findings suggest that these technologies have the potential to transform urban environments and enable the creation of more livable, sustainable, and efficient cities. However, significant challenges remain regarding data privacy, security, and ethical considerations, which must be addressed to realize the full potential of smart cities.

Abstract Due to the significance of trust in Social Internet of Things (SIoT)-based smart marketplaces, several research have focused on trust-related challenges. Trust is necessary for a smooth connection, secure systems, and dependable services during trade operations. Recent SIoT-based trust assessment approaches attempt to solve smart marketplace trust evaluation difficulties by using a variety of direct and indirect trust evaluation techniques and other local trust rating procedures. Nevertheless, these methodologies render trust assessment very sensitive to seller dishonesty, and a dishonest seller may influence local trust scores and at the same time pose a significant trust related threats in the system. In this article, a MarketTrust model is introduced, which is a blockchain-based method for assessing trust in an IoT-based smart marketplace. It has three parts: familiarity, personal interactions, and public perception. A conceptual model, assessment technique, and a global trust evaluation system for merging the three components of a trust value are presented and discussed. Several experiments were conducted to assess the model's security, viability, and efficacy. According to results, the MarketTrust model scored a 21.99% higher trust score and a 47.698% lower average latency than both benchmark models. Therefore, this illustrates that using the proposed framework, a potential buyer can efficiently choose a competent and trustworthy resource seller in a smart marketplace and significantly reduce malicious behavior.

Gastrointestinal diseases cause around two million deaths globally. Wireless capsule endoscopy is a recent advancement in medical imaging, but manual diagnosis is challenging due to the large number of images generated. This has led to research into computer-assisted methodologies for diagnosing these images. Endoscopy produces thousands of frames for each patient, making manual examination difficult, laborious, and error-prone. An automated approach is essential to speed up the diagnosis process, reduce costs, and potentially save lives. This study proposes transfer learning-based efficient deep learning methods for detecting gastrointestinal disorders from multiple modalities, aiming to detect gastrointestinal diseases with superior accuracy and reduce the efforts and costs of medical experts. The Kvasir eight-class dataset was used for the experiment, where endoscopic images were preprocessed and enriched with augmentation techniques. An EfficientNet model was optimized via transfer learning and fine tuning, and the model was compared to the most widely used pre-trained deep learning models. The model’s efficacy was tested on another independent endoscopic dataset to prove its robustness and reliability.

The main cause of stroke is the unexpected blockage of blood flow to the brain. The brain cells die if blood is not supplied to them, resulting in body disability. The timely identification of medical conditions ensures patients receive the necessary treatments and assistance. This early diagnosis plays a crucial role in managing symptoms effectively and enhancing the overall quality of life for individuals affected by the stroke. The research proposed an ensemble machine learning (ML) model that predicts brain stroke while reducing parameters and computational complexity. The dataset was obtained from an open-source website Kaggle and the total number of participants is 3,254. However, this dataset needs a significant class imbalance problem. To address this issue, we utilized Synthetic Minority Over-sampling Technique (SMOTE) and Adaptive Synthetic Sampling (ADAYSN), a technique for oversampling issues. The primary focus of this study centers around developing a stacking and voting approach that exhibits exceptional performance. We propose a stacking ensemble classifier that is more accurate and effective in predicting stroke disease in order to improve the classifier’s performance and minimize overfitting problems. To create a final stronger classifier, the study used three tree-based ML classifiers. Hyperparameters are used to train and fine-tune the random forest (RF), decision tree (DT), and extra tree classifier (ETC), after which they were combined using a stacking classifier and a k-fold cross-validation technique. The effectiveness of this method is verified through the utilization of metrics such as accuracy, precision, recall, and F1-score. In addition, we utilized nine ML classifiers with Hyper-parameter tuning to predict the stroke and compare the effectiveness of Proposed approach with these classifiers. The experimental outcomes demonstrated the superior performance of the stacking classification method compared to other approaches. The stacking method achieved a remarkable accuracy of 100% as well as exceptional F1-score, precision, and recall score. The proposed approach demonstrates a higher rate of accurate predictions compared to previous techniques.

In today’s world, services are improved and advanced in every field of life. Especially in the health sector, information technology (IT) plays a vigorous role in electronic health (e-health). To achieve benefits from e-health, its cloud-based implementation is necessary. With this environment’s multiple benefits, privacy and security loopholes exist. As the number of users grows, the Electronic Healthcare System’s (EHS) response time becomes slower. This study presented a trust mechanism for access control (AC) known as role-based access control (RBAC) to address this issue. This method observes the user’s behavior and assigns roles based on it. The AC module has been implemented using SQL Server, and an administrator develops controls for users with roles and access to multiple EHS modules. To validate the user’s trust value, A.net-based framework has been introduced. The framework of e-health proposed in this research ensures that users can protect their data from intruders and other security threats.

The exponential increase in textual unstructured digital data creates significant demand for advanced and smart stemming systems. As a preprocessing stage, stemming is applied in various research fields such as information retrieval (IR), domain vocabulary analysis, and feature reduction in many natural language processing (NLP). Text stemming (TS), an important step, can significantly improve performance in such systems. Text-stemming methods developed till now could be better in their results and can produce errors of different types leading to degraded performance of the applications in which these are used. This work presents a systematic study with an in-depth review of selected stemming works published from 1968 to 2023. The work presents a multidimensional review of studied stemming algorithms i.e., methodology, data source, performance, and evaluation methods. For this study, we have chosen different stemmers, which can be categorized as 1) linguistic knowledge-based, 2) statistical, 3) corpus-based, 4) context-sensitive, and 5) hybrid stemmers. The study shows that linguistic knowledge-based stemming techniques were widely used for highly inflected languages (such as Arabic, Hindi, and Urdu) and have reported higher accuracy than other techniques. We compare and analyze the performance of various state-of-the-art TS approaches, including their issues and challenges, which are summarized as research gaps. This work also analyzes different NLP applications utilizing stemming methods. At the end, we list the future work directions for interested researchers.

In the internet of things (IoT) networks, machine learning (ML) is significantly used for malware and adversary detection. Recently, research has shown that adversarial attacks have put ML-based models at risk. This problem is exacerbated in an IoT environment because of the absence of adequate security measures. Consequently, it is crucial to evaluate the strength of such malware detectors using powerful adversarial samples. The existing adversarial sample generation strategies either rely on high-level image features or an unfiltered feature set, making it challenging to determine which feature modifications are crucial in evading malware detection systems, without compromising the malware functionality. This encourages us to propose an evasion framework named IF-MalEvade, based on Generative Adversarial Network (GAN) and Deep Reinforcement Learning (DRL) that effectively generates fully-working, malware samples with several effective perturbations such as header Section manipulation and benign bytes insertion. The DRL framework selects a few suitable action sequences to change malicious samples, thus allowing our malware samples to bypass various black-box ML based malware detectors and the detection search engines of VirusTotal, while maintaining the executability and malicious behavior of the original malware samples. The neural networks of GAN take in the unfiltered feature set of malware dataset and using minimax objective function yields a set of useful features that are subsequently used by the DRL agent to make effective changes. Experimental results illustrated that by utilizing the influential features in sequence of transformations, the adversarial samples generated by our model outperformed the state-of-the-art evasion models with an impressive evasion rate. Additionally, the detection rate of well-known machine learning models was also brought down to up to 97%. Furthermore, when the machine learning models were retrained using adversarial samples, a 35% increase in detection accuracy was observed.

Particle swarm optimization (PSO) is an intelligent searching technique for solving complicated and multimodal design optimization’s problems. The classical PSO algorithm is more flexible and efficient because of its ability to solve a diverse range of complex and real-world issues. Moreover, the primary deficiency of this method that it trapped and stuck to local minima during the optimization of multimodal, complex and inverse objective function. We introduce a crossover and mutation vectors in the conventional PSO to solve this deficiency. The differential evolution strategies inspired the novel vectors. The central idea of the proposal is that, the novel global best particle is updated through a mutation vector and crossover vector. The introduction of the global best particle maintains the swarm diversity at the final steps of the evolution process. Also, we designed a novel strategy for the control parameter, which will maintain a decent alignment of the candidates between the global and local searches. The performance evaluation table and trajectory curves illustrate that our proposed approach is the best compared to other methods.

Diabetic Retinopathy (DR) is one of the leading causes of blindness and vision loss worldwide. According to the International Diabetes Federation (IDF), approximately one-third of individuals with diabetes, equivalent to 32.2%, are affected by some form of DR. Due to the uneven data distribution, intra-class variance, and a dearth of ophthalmologists, DR diagnosis is considered challenging. In recent years, Convolutional Neural Networks (CNN) and supervised learning techniques have been potentially useful in computer vision applications. However, unsupervised CNN has received less attention. Moreover, it is more manageable to use synthetic images for model training with the recent advancements in graphics. Therefore, the proposed method combines the actual and augmented views using the Deep Convolutional Generative Adversarial Network (DCGAN) algorithm. The generated views are implemented to balance the minority class in the imbalanced dataset. Furthermore, a novel ensemble convolutional neural network algorithm named Different View Ensemble (DVE) that merges the weighted average prediction of CNN, CNN-i, and CNN+i algorithms has been proposed. The proposed algorithm is evaluated on the DDR and EyePACS datasets, and its performance is compared with K-Means, Fuzzy C-Means (FCM), and Autoencoder-based Deep Embedded Clustering Techniques (DEC). The results demonstrate the superiority of the proposed algorithm, achieving an accuracy rate of 97.4%, specificity of 99.6%, and sensitivity of 92.3%. The promising results underscore the potential impact of this methodology in enhancing the accuracy and reliability of automated diagnostic systems in the field of ophthalmology. Notably, the evaluation considers imbalanced data and a DCGAN-balanced dataset, where the proposed approach exhibits even better performance with balanced classes.

Medical images are helpful for the diagnosis, treatment, and evaluation of diseases. Precise medical image segmentation improves diagnosis and decision-making, aiding intelligent medical services for better disease management and recovery. Due to the unique nature of medical images, image segmentation algorithms based on deep learning face problems such as sample imbalance, edge blur, false positives, and false negatives. In view of these problems, researchers primarily improve the network structure but rarely improve from the unstructured aspect. The paper tackles these challenges, accentuating the limitations of deep convolutional neural network-based methods and proposing solutions to reduce annotation costs, particularly in complex images, and introduces the improvement strategies to solve the problems of sample imbalance, edge blur, false positives, and false negatives. Additionally, the article introduces the latest deep learning-based applications in medical image analysis, covering segmentation, image acquisition, enhancement, registration, and classification. Moreover, the article provides an overview of four cutting-edge deep learning models, namely convolutional neural network (CNN), deep belief network (DBN), stacked autoencoder (SAE), and recurrent neural network (RNN). The study selection involved searching benchmark academic databases, collecting relevant literature and appropriate indicator for analysis, emphasizing DL-based segmentation and classification approaches, and evaluating performance metrics. The research highlights clinicians’ and scholars’ obstacles in developing an efficient and accurate malignancy prognostic framework based on state-of-the-art deep-learning algorithms. Furthermore, future perspectives are explored to overcome challenges and advance the field of medical image analysis.

The echo state network (ESN) is a cutting-edge reservoir computing technique designed to handle time-dependent data, making it highly effective for addressing time series prediction tasks. ESN inherits the more precise design of standard neural networks and the relatively simple learning process and has a strong computing capacity for solving nonlinear problems. It can disseminate low-dimensional information cues to high-dimensional areas enabling extracting data. However, this study has proven that not all reservoir output dimensions directly impact model generalization. This study desires to enhance the ESN model’s generalization abilities by decreasing the redundant reservoir output feature. A remarkable hybrid model is proposed that optimizes the ESN output association through feature selection. This model is called the binary improved gravitational search algorithm (BIGSA) echo state network (BIGSA-ESN). BIGSA’s feature selection approach complements the ESN output connection architecture. In this study, evaluation was performed using root mean square error (RMSE). The experimental findings on the Lorenz and Mackey-Glass benchmark time-series datasets demonstrate that the proposed technique outperforms conventional evolutionary methods. Moreover, empirical findings on predicting a significant water quality parameter from the wastewater treatment process (WWTP) dataset demonstrate that the proposed ensemble of BIGSA models performs very well in real-world scenarios.

Men’s fertility depends on their bodies making normal sperm and delivering them. Semen analysis has been the test of choice for assessing the male partner in an infertile couple using a single threshold value to distinguish ‘abnormal’ and ‘normal’ parameters. In the semen analysis process, rA©gime issues might affect the semen morphology, quality, and spermatozoa, and also reduce the risks of fertility due to food regimen including glycemic content and limited intake of nutrients. Determination of the connotation between adjustable rA©gime and semen morphology is a complex task to determine fertility. The goal is this study is the prediction of men’s fertility rate to analyze the connection between spermatozoa and the level of lifeblood. Impaired semen parameters alone cannot be used to predict fertility more accurately. Some factors might affect the spermatozoa, like impairing sperm function, morphology, and sustainability, and can reduce the men’s fertility rate. In this article, deep learning on convolutional neural network (DLNN) is used to predict the men’s fertility rate more quickly, accurately, and consistently from different age spam of men between 18–50 years old. The convolutional neural network performs the segmentation of sperm heads, while the deep learning algorithm allows us to calculate the movement speed of sperm heads. After the application of DLNN, we have achieved semen prediction 80.952% and sperm concentration 85.714% accuracy of sperm head detection on human spermatozoa sperm samples. The results of the experiments presented below will show the applicability of the proposed method to be used in automated artificial insemination workflow.

Around the world, brain tumors are becoming the leading cause of mortality. The inability to undertake a timely tumor diagnosis is the primary cause of this pandemic. Brain cancer diagnosis is a crucial procedure that relies on the expertise and experience of the doctor. Radiologists must use an automated tumor classification model to find brain cancers. The current model’s accuracy has to be improved to get suitable therapies. Radiologists can consult various computer-aided diagnostic (CAD) models in the literature on medical imaging to assist them with their patients. Previous research has widely used CNN models for tumor detection and classification, which typically require large datasets. This research proposed the Caps-VGGNet hybrid model, which integrates the CapsNet model with the VGGNet model by adding the layers of VGGNet. The presented model addresses the challenge of requiring large datasets by automatically extracting and classifying features. The suggested algorithm’s effectiveness was assessed using the Brats-2020 and Brats-2019 dataset, which contains high-quality images of brain tumors. Compared to other conventional and hybrid models, the empirical outcomes of the suggested model indicate that it exhibited the highest level of effectiveness and superior efficacy in terms of accuracy, specificity, and sensitivity. Specifically, the presented hybrid model attained an accuracy of 0.99, a specificity of 0.99, and a sensitivity of 0.98 on the Brats20 dataset.

Osteoarthritis is a deteriorating joint disease affecting millions worldwide. Osteoarthritis is a chronic condition that develops over time due to joint wear and tears. The degeneration of joint cartilage is the underlying cause of osteoarthritis, resulting in bone-to-bone contact and contributing to stiffness, discomfort, and restricted movement. People with osteoarthritis struggle to perform simple tasks such as walking, standing, or climbing stairs. Moreover, osteoarthritis can also cause psychological distress, including depression and anxiety, due to the chronic pain and disability associated with the condition. Improving the quality of life requires the development of efficient methods for early detection. Our study aims to create a model that can effectively diagnose osteoarthritis in knee X-ray images at an early stage. The advanced deep learning-based Convolutional Neural Network (CNN) and several machine learning-based techniques are applied in comparison. A novel transfer learning-based feature engineering technique CRK (CNN Random forest K-neighbors) is proposed to detect osteoarthritis with high performance. Using a 2D-CNN, the proposed CRK smartly extracts the spatial features from the X-ray images. The spatial features are input to the random forest and k-neighbors techniques, creating a probabilistic feature set. The probabilistic feature set is utilized to build the applied machine learning-based techniques. Extensive study experiments demonstrate that the proposed model outperformed with a 99% accuracy score for predicting osteoarthritis. The performance of each applied model is validated using hyperparameter optimization and k-fold-based cross-validation. The proposed study has the potential to revolutionize the prediction of osteoarthritis from X-ray images with high-performance scores.

Color information is useless for distinguishing significant edges and features in numerous applications. In image processing, a gray image discards much-unrequired data in a color image. The primary drawback of colour-to-grey conversion is eliminating the visually significant image pixels. A current proposal is a novel approach for transforming an RGB image into a grayscale image based on singular value decomposition (SVD). A specific factor magnifies one of the color channels (Red, Green, and Blue). A vector of three values (Red, Green, Blue) of each pixel in an image is decomposed using SVD into three matrices. The norm of the diagonal matrix was determined and then divided by a specific factor to obtain the grey value of the corresponding pixel. The contribution of the proposed method gives the user high flexibility to produce many versions of gray images with varying contrasts, which is very helpful in many applications. Furthermore, SVD allows for image reconstruction by combining the weighting of each channel with the singular value matrix. This results in a grayscale image that more accurately captures the actual intensity values of the image and preserves more color information than traditional grayscale conversion methods, resulting in loss of color information. The proposed method was compared with a similar method (converting the color image into grayscale) and was found to be the most efficient.

Deep learning has significantly aided current advancements in artificial intelligence. Deep learning techniques have significantly outperformed more than typical machine learning approaches, in various fields like Computer Vision, Natural Language Processing (NLP), Robotics Science, and Human-Computer Interaction (HCI). Deep learning models are ineffective in outlining their fundamental mechanism. That’s the reason the deep learning model mainly consider as Black-Box. To establish confidence and responsibility, deep learning applications need to explain the model’s decision in addition to the prediction of results. The explainable AI (XAI) research has created methods that offer these interpretations for already trained neural networks. It’s highly recommended for computer vision tasks relevant to medical science, defense system, and many more. The proposed study is associated with XAI for Sign Language Recognition. The methodology uses an attention-based ensemble learning approach to create a prediction model more accurate. The proposed methodology used ResNet50 with the Self Attention model to design ensemble learning architecture. The proposed ensemble learning approach has achieved remarkable accuracy at 98.20%. In interpreting ensemble learning prediction, the author has proposed SignExplainer to explain the relevancy (in percentage) of predicted results. SignExplainer has illustrated excellent results, compared to other conventional Explainable AI models reported in state of the art.

Due to their vast size, the growing number of digital images found in personal archives and on websites has become unmanageable, making it challenging to retrieve images from these large databases accurately. While these collections are popular due to their convenience, they often need to be equipped with proper indexing information, making it difficult for users to find what they need. One of the most significant challenges in computer vision and multimedia is image annotation, which involves labeling images with descriptive keywords. However, computers need to possess the capability to understand the essence of images in the same way that humans do, and people can only identify images based on their visual attributes rather than their deeper semantic meaning. Therefore, image annotation requires keywords to effectively communicate the contents of an image to a computer system. However, raw pixels in an image need to provide more information to generate semantic concepts, making image annotation a complex task. Unlike text annotation, where the dictionary linking words to semantics is well established, image annotation lacks a clear definition of “words” or “sentences” that can be associated with the meaning of the image, known as the semantic gap. To address this challenge, this study aimed to characterize image content meaningfully to make information retrieval easier. An improved automatic image annotation (AIA) system was proposed to bridge the semantic gap between low-level computer features and human interpretation of images by assigning one or multiple labels to images. The proposed AIA system can convert raw image pixels into semantic-level concepts, providing a clearer representation of the image content. The study combined the ResNet50 and slantlet transform with word2vec and principal component analysis with t-distributed stochastic neighbor embedding to balance precision and recall. This allowed the researchers to determine the optimal model for the proposed ResNet50-SLT AIA framework. A Word2vec model with ResNet50-SLT was used with principal component analysis and t-distributed stochastic neighbor embedding to improve IA prediction accuracy. The distributed representation approach involved encoding and storing information about image features. The proposed AIA system utilized seq2seq to generate sentences depending on feature vectors. The system was implemented on the most popular datasets (Flickr8k, Corel-5k, ESP-Game). The results showed that the newly developed AIA scheme overcame the computational time complexity associated with most existing image annotation models during the training phase for large datasets. The performance evaluation of the AIA scheme showed its excellent flexibility of annotation, improved accuracy, and reduced computational costs, thus outperforming the existing state-of-the-art methods. In conclusion, this AIA framework can provide immense benefits in accurately selecting and extracting image features and easily retrieving images from large databases. The extracted features can effectively be used to represent the image, thus accelerating the annotation process and minimizing the computational complexity.

Human body skeleton, acting as a spatiotemporal graph, is increasing attentions of researchers to adopt graph convolutional networks (GCN) to mine the discriminative features from skeleton joints. However, one of GCN’s flaws is its inability to handle long-distance reliance between joints. In this regard, graph attention network (GAT) was recently suggested, which combines graph convolutions with a self-attention mechanism to extract the most informative joint of a human skeleton and increase the model accuracy. However, GAT can compute only static attention: for each query node, the attention rank is same which severely hurts and limits the expressivity of an attention mechanism. In this work, we present a spatial-temporal dynamic graph attention network (ST-DGAT) to learn the spatial-temporal patterns of skeleton sequences. For dynamic graph attention, we tweak the order of weighted vector operations in GAT, our approach achieves a global approximate attention function, making it strictly superior to GAT. Experiments show that by fixing the order of internal operation of GAT the proposed model achieved better action classification results while maintaining the same computing cost as GAT. The proposed framework has been evaluated on well-known publicly available large-scale datasets NTU60, NTU120, and Kinetics-400, which notably outperforms state-of-the-art (SOTA) results with an accuracy of 96.4%, 88.2%, and 61.0%, respectively.

Sign language is the most common form of communication for the hearing impaired. To bridge the communication gap with such impaired people, a normal people should be able to recognize the signs. Therefore, it is necessary to introduce a sign language recognition system to assist such impaired people. This paper proposes the Transformer Encoder as a useful tool for sign language recognition. For the recognition of static Indian signs, the authors have implemented a vision transformer. To recognize static Indian sign language, proposed methodology archives noticeable performance over other state-of-the-art convolution architecture. The suggested methodology divides the sign into a series of positional embedding patches, which are then sent to a transformer block with four self-attention layers and a multilayer perceptron network. Experimental results show satisfactory identification of gestures under various augmentation methods. Moreover, the proposed approach only requires a very small number of training epochs to achieve 99.29 percent accuracy.

Being the only nonlinear component in many cryptosystems, an S-box is an integral part of modern symmetric ciphering techniques that creates randomness and increases confidentiality at the substitution stage of the encryption. The ability to construct a cryptographically strong S-box solely depends on its construction scheme. The primary purpose of an S-box in encryption standards is to establish confusion between the $m$ -bit input into the $n$ -bit output (both $m, n >= 2$ ). This article proposed a robust way to construct S-boxes based on the Gravesian octonion integers. We chunk the paper into threefold: firstly, a comprehensive technique for constructing S-box using affine mapping is described. The presented work is developed in such a way that for every valid input, it generates two S-boxes. Secondly, the strength of the newly generated S-box is evaluated by passing through a rigorous security analysis. Finally, a thorough comparison of the newly developed method with some well-known existing schemes is conducted. We mainly targeted some elliptic curve-based S-boxes in comparison by taking the same parameters in our scheme. The computational results and performance analysis reveal that the propose algorithm can construct a large number of distinct S-boxes that are cryptographically secured and create high resistance against various cryptanalysis attacks.

Abstract Soil salinity accumulates a high concentration of salts in soils that interfere with normal plant growth. Early detection and quantification of soil salinity are essential to effectively deal with soil salinity in agriculture. Soil salinity quantification and mapping at the irrigation scheme level are vital to evaluating saline soil's reclamation activity. Existing solutions of salinity mapping are costly, time-consuming, and inadequate for applications at the irrigation scheme level. Internet of Things (IoT) assisted salinity mapping at the irrigation scheme level is proposed to quantify and map the soil salinity in agriculture. The proposed IoT-assisted salinity mapping characterizes the soil salinity in terms of Electric Conductivity, pH, and Total Dissolved Salts. The proposed IoT-assisted salinity mapping effectively observes impacts of reclamation activities in saline soil by frequent observation of soil salinity cost-effectively. The accuracy of proposed IoT-assisted salinity mapping is evaluated against the standard method of salinity measurements. The proposed IoT-assisted salinity mapping is cost-effective, and portable, which is very useful for site-specific treatments and soil zones management in saline soils.

Diabetic Retinopathy (DR) is a prevalent acute stage of diabetes mellitus that causes vision-effecting abnormalities on the retina. This will cause blindness if not identified early. Because DR not an irreversible procedure, and only vision is preserved via care. Consequently, Early diagnosis and care with DR will significantly minimize the chance of vision loss. In modern ophthalmology, retinal image analysis has become a popular approach to disease diagnosis. The ophthalmologists and computerized systems extensively employ fundus angiography to detect DR-based clinical signs for early detection of DR. fundus photographs are commonly prone to low contrast, noise, and irregular illumination issues due to the complexity of imaging environments such as imaging variety of angles and light conditions. This research presents an Algorithm for improving the quality of images to strengthen the standard of color fundus images by reducing the noise and improving the contrast. The approach includes two main stages: cropping the images to remove insignificant content, then applying the shape crop and gaussian blurring for noise reduction and contrast improvement. The experimental results are evaluated using two standard datasets EyePACS and MESSIDOR. It’s clearly shown that the outcomes of feature extraction and classification of enhanced images is outperform the results without applying the enhancement approach. The improved algorithm is also tested in smart hospitals as an IoMT application.

An application or product is considered “usable” if it is pleasing, easy to use, and works as expected user interface. Most companies majorly focus on the application’s functional requirements but put minimal effort into user experience (usability). Consumers’ adaptation of these applications depends on the number of features and user interface. In this work, an android based application, “Houzcalls” is used for usability study using PACMAD usability model. This work is focused on the variations in PACMAD attributes based on the participants’ education and age. Participants are segregated into two major groups FG1 and FG2 based on their education. All the participants with more than 10 years of education are in FG1 while others in FG2. Each focal group is divided into four subgroups Under 25, 25-35, 36-45, and Over 45 based on their ages. The results have shown participants in FG1 have shown more Effectiveness, Efficiency, Satisfaction, Learnability, and Memorability. In contrast, they have committed fewer Errors and shown less Cognitive Load during usability testing as compared to FG2. These variations can also be seen age-wise as generally “Under 25 and 25-35” subgroups have shown better results than other subgroups. It is inferred from the study that application usability and acceptability can be increased by considering the general population during development which includes all groups of people based on education and age.

Every day, websites and personal archives generate an increasing number of photographs. The extent of these archives is unfathomable. The ease of usage of these enormous digital image collections contributes to their popularity. However, not all of these databases provide appropriate indexing data. As a result, it’s tough to find information that the user is interested in. Thus, in order to find information about an image, it is necessary to classify its content in a meaningful way. Image annotation is one of the most difficult issues in computer vision and multimedia research. The objective is to convert an image into a single or numerous labels. This necessitates a grasp of the visual content of an image. The necessity for unambiguous information to build semantic-level concepts from raw image pixels is one of the challenges of image annotation. Unlike text annotation, where a dictionary links words to their meaning, raw picture pixels are insufficient to construct semantic-level notions directly. A simple syntax, on the other hand, is well specified for combining letters to form words and words to form sentences. The automatic feature extraction for automatic annotation was the emphasis of this paper. And they employed a deep learning convolutional neural network to build and improve image coding and annotation capabilities. Performance of the suggested technique on the Corel-5K, ESP-Game, and IAPRTC-12 datasets. Finally, experimental findings on three data sets were used to demonstrate the usefulness of this model for image annotation.

Accurate fruit counting is one of the significant phenotypic traits for crucial fruit harvesting decision making. Existing approaches perform counting through detection or regression-based approaches. Detection of fruit instances is very challenging because of the very small fruit size compared to the whole size image of a tree. At the same time, regression-based counting techniques contributes impressive results but presents inaccurate results while number of instances increases. Moreover, most approaches lack scalability and are applicable only on one or two fruit types. This paper proposes a fruit counting mechanism that combines loose segmentation and regression counting that works on six fruit types: Apple, Orange, Tomato, Peach, Pomegranate and Almond. Through relaxed segmentation, fruit clusters are segmented to extract the small image regions which contain the small cluster of fruits. Extracted regions are forwarded for the regression counting of fruits. Relaxed segmentation is achieved through a state-of-the-art deconvolutional network, while modified Inception Residual Networks (ResNet) based nonlinear regression module is proposed for fruit counting. For segmentation, 4,820 original images, including corresponding mask images, of all six fruit types are augmented to 32,412 images through different augmentation techniques, while 21,450 extracted patches are augmented to 89,120 images used for the regression module training. The proposed approach attained a counting accuracy of 94.71% for individual fruit types higher than techniques reported in literature.

Alzheimer’s disease is an incurable neurological disorder that leads to a gradual decline in cognitive abilities, but early detection can significantly mitigate symptoms. The automatic diagnosis of Alzheimer’s disease is more important due to the shortage of expert medical staff, because it reduces the burden on medical staff and enhances the results of diagnosis. A detailed analysis of specific brain disorder tissues is required to accurately diagnose the disease via segmented magnetic resonance imaging (MRI). Several studies have used the traditional machine-learning approaches to diagnose the disease from MRI, but manual extracted features are more complex, time-consuming, and require a huge amount of involvement from expert medical staff. The traditional approach does not provide an accurate diagnosis. Deep learning has automatic extraction features and optimizes the training process. The Magnetic Resonance Imaging (MRI) Alzheimer’s disease dataset consists of four classes: mild demented (896 images), moderate demented (64 images), non-demented (3200 images), and very mild demented (2240 images). The dataset is highly imbalanced. Therefore, we used the adaptive synthetic oversampling technique to address this issue. After applying this technique, the dataset was balanced. The ensemble of VGG16 and EfficientNet was used to detect Alzheimer’s disease on both imbalanced and balanced datasets to validate the performance of the models. The proposed method combined the predictions of multiple models to make an ensemble model that learned complex and nuanced patterns from the data. The input and output of both models were concatenated to make an ensemble model and then added to other layers to make a more robust model. In this study, we proposed an ensemble of EfficientNet-B2 and VGG-16 to diagnose the disease at an early stage with the highest accuracy. Experiments were performed on two publicly available datasets. The experimental results showed that the proposed method achieved 97.35% accuracy and 99.64% AUC for multiclass datasets and 97.09% accuracy and 99.59% AUC for binary-class datasets. We evaluated that the proposed method was extremely efficient and provided superior performance on both datasets as compared to previous methods.

Abstract Agriculture plays a critical role in the economy of several countries, by providing the main sources of income, employment, and food to their rural population. However, in recent years, it has been observed that plants and fruits are widely damaged by different diseases which cause a huge loss to the farmers, although this loss can be minimized by detecting plants’ diseases at their earlier stages using pattern recognition (PR) and machine learning (ML) techniques. In this article, an automated system is proposed for the identification and recognition of fruit diseases. Our approach is distinctive in a way, it overcomes the challenges like convex edges, inconsistency between colors, irregularity, visibility, scale, and origin. The proposed approach incorporates five primary steps including preprocessing,Standard instruction requires city and country for affiliations. Hence, please check if the provided information for each affiliation with missing data is correct and amend if deemed necessary. disease identification through segmentation, feature extraction and fusion, feature selection, and classification. The infection regions are extracted using the proposed adaptive and quartile deviation-based segmentation approach and fused resultant binary images by employing the weighted coefficient of correlation (CoC). Then the most appropriate features are selected using a novel framework of entropy and rank-based correlation (EaRbC). Finally, selected features are classified using multi-class support vector machine (MC-SCM). A PlantVillage dataset is utilized for the evaluation of the proposed system to achieving an average segmentation and classification accuracy of 93.74% and 97.7%, respectively. From the set of statistical measure, we sincerely believe that our proposed method outperforms existing method with greater accuracy.

Text to face generation is a sub-domain of text to image synthesis. It has a huge impact on new research areas along with the wide range of applications in the public safety domain. Due to the lack of dataset, the research work focused on the text to face generation is very limited. Most of the work for text to face generation until now is based on the partially trained generative adversarial networks, in which the pre-trained text encoder has been used to extract the semantic features of the input sentence. Later, these semantic features have been utilized to train the image decoder. In this research work, we propose a fully trained generative adversarial network to generate realistic and natural images. The proposed work trained the text encoder as well as the image decoder at the same time to generate more accurate and efficient results. In addition to the proposed methodology, another contribution is to generate the dataset by the amalgamation of LFW, CelebA and locally prepared dataset. The dataset has also been labeled according to our defined classes. Through performing different kinds of experiments, it has been proved that our proposed fully trained GAN outperformed by generating good quality images by the input sentence. Moreover, the visual results have also strengthened our experiments by generating the face images according to the given query.

Prostate cancer (PCa) is a severe type of cancer and causes major deaths among men due to its poor diagnostic system. The images obtained from patients with carcinoma consist of complex and necessary features that cannot be extracted readily by traditional diagnostic techniques. This research employed deep learning long short-term memory (LSTM) and Residual Net (ResNet - 101), independent of hand-crafted features, and is fine-tuned. The results were compared with hand-crafted features such as texture, morphology, and gray level co-occurrence matrix (GLCM) using non-deep learning classifiers such as support vector machine (SVM) Gaussian Kernel, k-nearest neighbor-Cosine (KNN - Cosine), kernel naive Bayes, decision tree (DT) and RUSBoost tree. This study reduces the features of carcinoma images, employed machine learning and deep learning approaches. For validation of training and testing data, a jack-knife ten-fold cross-validation method was used. The performance was measured using a confusion matrix such as sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy (AC), Mathews Correlation Coefficient (MCC), and area under the curve (AUC). The most remarkable performance was obtained using non-deep learning methods with GLCM features using KNN-Cosine with sensitivity (98.00%), specificity (99.25%), PPV (98.99%), NPV (99.11%), accuracy (99.07%), and AUC (0.998). The LSTM deep learning method yields performance with sensitivity (98.33%), specificity (100%), PPV (100%), NPV (99.26%), accuracy (99.48%), MCC (0.9879) and AUC (0.9999), where using Deep learning method ResNet - 101, we obtained (100%) Accuracy and AUC (1) for Kernel Naive Bayes, SVM Gaussian and RUSBoost Tree. The results show that ResNet - 101 deep learning outperformed than non-deep learning methods and LSTM. Thus, the deep learning method ResNet - 101 could be used as a better predictor for the detection of prostate cancer.

Recently, much attention has been given to image annotation due to the massive increase in image data volume. One of the image retrieval methods which guarantees the retrieval of images in the same way as texts are automatic image annotation (AIA). Consequently, numerous studies have been conducted on AIA, particularly on the classification-based and probabilistic modeling techniques. Several image annotation techniques that performed reasonably on standard datasets have been developed over the last decade. In this paper, a review of the image annotation method was conducted, focusing more on deep learning models. Automatic image annotation (AIA) methods were also classified into five categories, including i) Convolutional Neural Network (CNN) based on AIA, ii) Recurrent Neural Network (RNN) based on AIA, iii) Deep Neural Networks (DNN) based on AIA, iv) Long-Short-Term Memory (LSTM) based on AIA, and v) Stacked auto-encoder (SAE) based on AIA. An assessment of the five varieties of AIA methods was also offered based on their principal notion, feature mining technique, explanation precision, computational density, and examined aggregated data. Moreover, the evaluation metrics used to evaluate AIA methods were reviewed and discussed. The need for careful consideration of methods throughout the improvement of novel procedures and datasets for image annotation assignment was highly demanded. From the analysis of the achievements so far, it is certain that more attention should be paid to automatic image annotation.

The protection of digital content is increasingly becoming a significant issue for researchers and engineers. In this context, nonlinear dynamic systems play a vital role in information security through their chaotic behavior and susceptibility to initial conditions. This research presents a 3D chaotic map-based symmetric algorithm for multiple images to improve encryption efficiency and encourage secure transmission. The proposed scheme comprises the following four modules: the combination (the images are combined into a single image by merging their color channels); the permutation (using the suggested 3D chaotic map); the S-box generation; and the substitution through the AES substitution method. The proposed algorithm’s encryption strength was determined through Entropy, Correlation coefficient, NPCR, and UACI analyses, which were then compared to the past techniques. Furthermore, the proposed method is assessed in terms of its computation time. Results demonstrate that it is highly efficient and secure for real-time communication.

Corona Virus is a pandemic, and the whole world is affected due to it. Apart from the vaccine, the only cure for this drastic disease is to follow the rules and regulations that avoid further spread. There are different mechanisms like (Social Distancing, Mask Detection, Human occupancy etc.) through which we can able to stop the spread of the coronavirus. In this paper, we proposed hotspot zone detection using the computer vision techniques of deep learning. We have defined the hotspot area as the particular region on which the person touches more than some specified threshold. We further mark that area to some specific color to help the authority take necessary action and disinfect that particular place. To implement this algorithm, we have utilized the human-object interaction concept. We have extracted the dataset of person classes from the publicly available dataset for the person detection and the self-generated dataset to train the algorithm. Different experiments on object detection algorithms (YOLO-v3, Faster RCNN, SSD) for person detection have been performed in this work. We achieved the maximum accuracy in real-time on the YOLO-v3 for person detection. Whereas we have marked the specific area using the template matching algorithm of computer vision techniques. Our proposed algorithm detects the persons and extracts the region of interest points on which the user draws the rectangle. Then we find the intersection over union ratio between the detected person and the region of interest of the marked area to make the decision. We have achieved 88.72% accuracy on person detection in the local environment. Whereas, for the whole system of human-object interaction for detecting the hotspot area zone, we have achieved 86.7% accuracy using the confusion matrix.

Nowadays, modeling and forecasting electricity spot prices are challenging due to their specific features, including multiple seasonalities, calendar effects, and extreme values (also known as jumps, spikes, or outliers). This study aims to provide a comprehensive analysis of electricity price forecasting by comparing several outlier filtering techniques followed by various modeling frameworks. To this end, extreme values are first treated with five different filtering techniques and are then replaced by four different outlier replacement approaches. Next, the spikes-free series is divided into deterministic and stochastic components. The deterministic component includes long-term trend, yearly and weekly seasonalities, and bank holidays and is estimated through parametric and nonparametric approaches. On the other hand, the stochastic component accounts for the short-run dynamics of the price time series and is modeled using different univariate and multivariate models. The one-day-ahead out-of-sample forecast results for the Italian Power Exchange (IPEX), obtained for a whole year, suggest that the outliers pre-filtering give a high accuracy gain. In addition, multivariate modeling for the stochastic component outperforms univariate models.