Abstract The forecasting of multi-hazards is a vital, though underinvestigated, area of disaster risk management. The traditional studies have mainly focused on single-hazard forecasting, thus leaving its utility in real-world and realistic scenarios. This study, in turn, presents a spatio-temporal multi-label classification model, a framework designed expressly to capture the complex interrelationships between a range of hazards. The methodological framework used disaster occurrence data from the Open Federal Emergency Management Agency (OpenFEMA) database and converted the raw records of disasters into a multi-label dataset. Pressure-level reanalysis data is extracted from Climate Data Store (CDS) based on the multi-hazard event. Spatial data is extracted in 25 $$\times$$ 59 grid format in different temporal dependencies (12 h, 8 h, 6 h) at the 850 hPa pressure level. The model architecture combines convolutional neural networks (CNNs) with spatial attention mechanisms and gated recurrent units (GRUs) that model the temporal sequences. This combination enables multi-hazard predictions by utilizing the spatial and temporal data. Experimental analysis reveals that the proposed model outperformed the baseline variants, i.e., 2D CNN, Convolutional Long Short-Term Memory (ConvLSTM), and Convolutional Gated Recurrent Unit (ConvGRU) without attention. The proposed model achieved per-class accuracy up to 0.8868, the subset accuracy is 0.55, and the Hamming loss up to 0.127, which are 3.88%, 13.59% and 21.12% performance improvements over the baseline models respectively. In addition, the use of various lead times and the fusion of multiple lead times (12 h+8 h+6 h) significantly improves the predictive capability. The proposed framework has high potential for disaster preparedness and early warning systems in the real world. It proposes a flexible and efficient method of dealing with the growing complexity of multi-hazard environments.

Abstract In recent decades, fog computing has contributed significantly to the expansion of smart cities. It generated numerous real‐time data and coped with time‐constraint applications. They use sensors, physical objects, and network standards to monitor health imaging, traffic surveillance, industrial management, and so forth. Interactive applications have been proposed for the Internet of Things (IoT) to control wireless channels and improve communication. However, most of the existing lack of handing network interference and a reliable monitoring process. Moreover, many solutions are vulnerable to external threats, resulting in inconsistent and untrustworthy information for end users. Thus, this article proposes a framework that considers possible shortest paths to provide the most reliable and low‐latency healthcare decision system using Q‐learning. In addition, fog devices offer a trusted transmission interference system and are kept secure. The proposed framework is specially designed for rapid real‐time medical data processing while enforcing robust security throughout the IoT‐based transmission process. To identify the health sensors in pairwise objects with the initial computing cost, the proposed framework applies graph theory. It also extracts the most effective and least loaded communication edges by examining the behaviour of devices. Moreover, the identities of devices are verified using lightweight timestamps and secret information, accordingly, it decreases the privacy threats.

The prompt expansion of Internet of Things (IoT) devices necessitates advanced security frameworks to protect data integrity, confidentiality, and availability in resource-constrained environments. Traditional security solutions are often resource-intensive for IoT devices with limited computational power and energy resources. This study addresses these inadequacies by proposing a novel approach formulated to such constraints. This study propose the trust-enhanced lightweight security framework (TELSF), integrating two novel components: 1) the adaptive lightweight encryption algorithm (ALEA) and 2) the trust-aware data protection model (TADPM). ALEA employs dynamic key generation through a lightweight hash function, ensuring unique and regularly updated encryption keys based on device context and behavior. TADPM enhances this framework by continuously assessing device trustworthiness through direct interactions, aggregated feedback from neighboring devices, and contextual parameters, such as location and device capabilities. Performance evaluations demonstrate that TELSF significantly enhances security and operational efficiency, reducing computational overhead by 18%, improving energy efficiency by 20%, and increasing data transmission security by 10% compared to existing solutions.

Ultrasound imaging continues to play a critical role in prenatal diagnostics, but accurate interpretation remains hindered by limited labeled data, inconsistent pseudo label quality, and real-time processing constraints in Internet of Medical Things (IoMT) environments. Existing semi-supervised learning (SSL) frameworks fail to maintain reliable segmentation under these dynamic and resource-constrained conditions. This study proposes ADLIFT, a real-time SSL-based ultrasound processing framework designed to optimize diagnostic accuracy and computational efficiency. The approach integrates an Adaptive Dual-Layer Perception (ADLP) mechanism combining macro-level anatomical recognition with micro-level feature refinement, and a Dynamic Label Generation (DLG) module that iteratively improves pseudolabel reliability using confidence-driven feedback. Efficient Sparse Feature Extraction (ESFE) minimizes computational overhead by isolating high-activation regions, while the Temporal Contextualization Framework (TCF) ensures inter-frame consistency. Blockchain-enhanced edge computing supports secure and scalable IoMT deployment. Evaluations in HC18, FetalPlane18 and Kvasir-Segment datasets demonstrate precision of 93. 7%, decision stability of 92. 8%, interpretability index of 91. 5%, uncertainty handling efficiency of 89. 7%, trust reliability score of 95. 3%, and processing latency of 28.1 ms per frame.

Abstract The Internet of Health Things (IoHT) has grown in importance for developing medical applications with the support of wireless communication systems. IoHT is integrated with many sensors to capture the patients' records and transmits them to hospital centres for analysis and reporting. Controlling and managing health records has been addressed in several ways, however, it is noted that two key research problems for vital communication systems are reliability and reducing data loss. To enhance the sustainability of health applications and effectively use the network infrastructure when transferring sensitive data, this research provides a machine learning approach. Moreover, data collected from the IoHTs are protected and can be securely received for physical process in hospitals using authentication trees. Firstly, the undirected graphs are explored based on the multi‐parametric machine learning approach to minimize the computation overheads and traffic congestion. Secondly, it evaluates the nodes' level behaviour over the heterogeneous traffic load with efficient identification of redundant links. Finally, in‐depth analysis and simulation results have shown that the proposed protocol is more effective than existing approaches for data accuracy and security analysis.

By exploring the future network, smart technologies promote the development of cutting-edge industrial applications. Internet of Things (IoT) systems use sensing approaches to acquire data and control real-time processing and complex tasks. Several techniques have been proposed for coping with environmental behavior in industrial management and reducing the response in crucial circumstances. However, due to the unique and limited constraints of the industrial environment, managing data routing and sustainable development are recent research concerns. In addition, security is essential for industrial communication systems due to the probability of unauthorized access, thus trust level must be improved. The framework addresses real-world challenges in industrial networks by incorporating a lightweight data verification algorithm designed for green communication, reducing energy consumption while maintaining data integrity. First, predictive computing is implemented using ant colony optimization (ACO) based on real-time requirements and selects the dynamic and communication channels for data transmission across the industrial platform. Second, mobile sinks offer more authentic techniques for verifying sensor data and delivering it securely to the cloud servers. The framework was evaluated and validated in a simulation-based environment, revealing a considerable improvement in terms of network throughput, packet drop ratio, connectivity ratio, and network overhead over the existing approaches.

Current human being's lifestyle has caused / exacerbated many diseases. One of these diseases is cancer, and among all kinds of cancers like, brain pulmonary; lung cancer is fatal. The cancers could be detected early to save lives using Computer Aided Diagnosis (CAD) systems. CT scans medical images are one the best images in detecting these tumors in lungs that are especially accepted among doctors. However, the location, random shape of tumors, and poor quality of CT scan images are among the main challenges for physicians in identifying these tumors. Therefore, deep learning algorithms have been highly regarded by researchers. This paper proposed a new model for tumors and nodules segmentation in CT scans images based on convolution neural network (CNN) algorithm. The proposed model comprises preprocessing and postprocessing for fine segmentation of nodules. Filtering is used for image enhancement in preprocessing, and morphological operators are used for fine segmentation in post-processing. Finally, the active counter algorithm implementation exhibited tumors and nodules detection precisely. The sensitivity assessment and dice similarity criteria qualitatively measure the proposed model efficiency on the benchmark dataset. The obtained results with 98.33% accuracy 99.25% validity,98.18% dice similarity criterion show superiority of the proposed model.

In the recent era, the communication network with the support of many wireless technologies is giving benefits for remote access. Such a communication model increases the flexibility for data storage with the management of network resources efficiently with the integration of the Internet of Things (IoT). Although, the industrial Internet of Things (IIoT) has enabled the development of numerous machine learning-based solutions to provide real-time applications. However, most of the solutions are not prepared to cope with heterogeneous services and the huge amount of device data in the digital world. Furthermore, conducting financial transactions over the Internet raises several security issues. Such restrictions compromise the sensitive data of financial institutions and also degrade the trust of network users in the system. Thus, this article presents a secured blockchain model for high-performance computing in a big data environment, which aims to protect the business activities for financial interaction with intelligent services of software-defined network (SDN) architecture. First, to keep the security credentials, the SDN controller creates an association between the IoT devices and maintains local and global records. Second, the machine learning approach is explored using reliable and fault-tolerant methods to support the network scalability and extract the updated routing information for transmitting financial data. The proposed protocol also provides data integrity with a high level of network availability and copes with the financial security of big data by investigating cryptographic approaches. Our proposed protocol is tested using simulation, and various experiments are performed to show its efficacy in terms of network throughput, computing overhead, data delay, response time, and dropped packets as compared to tunicate swarm algorithm-based optimized routing mechanism (TORM) and RouteChain.

A stroke occurs when there is a sudden interruption of blood flow to a specific area of the brain. Brain cells perish and lose functionality when deprived of blood flow, depending on the specific region of the brain impacted. Early diagnosis of symptoms could provide useful insights for supporting a healthy life and predicting strokes. Considerable endeavours have been made to enhance stroke prevention and management due to the profound impact strokes have on populations. This paper presents a hybrid deep learning model to predict the risk of an early-stage brain stroke. To evaluate the model’s effectiveness, a benchmark dataset for stroke prediction was selected through the online Kaggle platform. The dataset was cleaned and normalised using a variety of preprocessing techniques in order to predict strokes. LSTM, RNN, CNN, and GRU models were among the models used in this study to perform classification tasks. The results demonstrated that CNN+GRU attained 98.78% accuracy and LSTM+RNN attained 98.23% accuracy rate. The empirical findings also exhibited that deep learning techniques were more effective than approaches reported in state of the art.

Abstract The automated behavior analysis is a significant problem in the examination. In the presence of invigilators and examiners, examinees use various unfair means to cheat the staff. Examinees may use different gestures like moving their head, eyes and using different suspicious objects (mobile phone, calculator) and so forth. This research is intended to automatically identify and distinguish examinees in live videos who cheat by their activities in examination hall during exams. Hence, we took the privilege of ensemble learning using deep learning‐based algorithms. The suspicious head movements and prohibited objects have been monitored using fine‐tuned Faster RCNN algorithm. At the same time, interactive use‐age of prohibited objects find out using Open‐Pose architecture. For this purpose, intersection over union (IOU) has been calculated between the region of interest (ROI) of detected objects and pose points of hands and face. To get the identity of a specific examinee, we stored the features of facial ROI after the last convolution layer and matched them at the time of testing for statistical analysis. We achieved state‐of‐the‐art results using standard evaluating measures. Moreover, for experiments, dataset of automatic examinee invigilation is available currently. So, this research also contributes toward the generation and annotation of the dataset.

The Internet of Things (IoT) connects heterogeneous sensors with dynamic networks to monitor smart communication and collect real-time data. Such systems are well adapted to satisfy the needs of smart cities and facilitate remote locations. Many cloud-based solutions for effective routing along with scalable data storage have been presented for constraint IoT systems. However, because of the unpredictable nature of mobile networks and communication links, most of the solutions may not be suitable for realistic applications and usually result in path failure with increasing resource utilization. Hence, data forwarding is only reliable and valuable if the proposed algorithms are trust aware with low overheads and consume balanced energy among nodes. Therefore, this article proposed a fault-tolerant supervised routing (Trust-FTSR) model for trust management in the IoT network, to improve trustworthiness and collaborative communication in smart cities. Each node evaluates the behavior of its neighbors and establishes a direct trust for a reliable and optimized network structure. In addition, using a supervised machine-learning technique, a fault-tolerant relaying system is provided without imposing additional overheads. Moreover, it removes the additional load in determining the optimal decision and training the IoT system to balance the network cost. In the end, a secure algorithm is proposed to ensure the privacy and authentication of the relaying system in the presence of critical attacks with secured keys. The proposed model is tested and its performance has significant improvement as compared to existing work.

Skin cancer is actually considered one of the most harmful human types of cancer. It exists in many types, but melanoma is the most severe. Early detection of melanoma cancer is valuable for the treatment of patients. For this function, computer vision plays a major role in medical imaging for the diagnosis of cancer. Using Jetson Nano, we established an image processing method for skin cancer detection at the initial stage in the proposed work. The proposed research framework consists of five phases: a collection of dermoscopic images, grayscale conversion of an image, area of interest segmentation, and noise removal. Finally, the lesion features are extracted from the lesion and categorized into three groups with the aid of ABCD rules: benign, suspicious and malignant. On PH2 and ISIC data sets of skin lesion images, experiments are performed. Compared to those published in state of the art for skin cancer identification, the proposed method has shown better results.

Mobile technology is very fast growing and incredible, yet there are not much technology development and improvement for Deaf‐mute peoples. Existing mobile applications use sign language as the only option for communication with them. Before our article, no such application (app) that uses the disrupted speech of Deaf‐mutes for the purpose of social connectivity exists in the mobile market. The proposed application, named as vocalizer to mute (V2M), uses automatic speech recognition (ASR) methodology to recognize the speech of Deaf‐mute and convert it into a recognizable form of speech for a normal person. In this work mel frequency cepstral coefficients (MFCC) based features are extracted for each training and testing sample of Deaf‐mute speech. The hidden Markov model toolkit (HTK) is used for the process of speech recognition. The application is also integrated with a 3D avatar for providing visualization support. The avatar is responsible for performing the sign language on behalf of a person with no awareness of Deaf‐mute culture. The prototype application was piloted in social welfare institute for Deaf‐mute children. Participants were 15 children aged between 7 and 13 years. The experimental results show the accuracy of the proposed application as 97.9%. The quantitative and qualitative analysis of results also revealed that face‐to‐face socialization of Deaf‐mute is improved by the intervention of mobile technology. The participants also suggested that the proposed mobile application can act as a voice for them and they can socialize with friends and family by using this app.

Abstract With ceaseless utilization of web and other online advances, it has turned out to be amazingly simple to imitate, discuss, and convey digital material. Subsequently, confirmation and copyright assurance issues have been emerged. Text is the most widely used media of communication on the web as compared with images and videos. The significant part of books, daily papers, websites, commercial, research papers, reports, and numerous different archives are just the plain text. Therefore, copyrights protection of plain text is a critical issue that could not be accepted at all. This paper presents a concise analysis on information hiding techniques and their consequences. Various watermarking systems properties are highlighted along with different types of attacks and their possible defenses. Additionally, the current applications for text document watermarking and the reasons for the problems of text document watermarking are described. Finally, attacks characteristics are discussed including types of attacks, volumes, and nature on text. Current watermarking approaches are reviewed in depth, and consequently, their advantages and weaknesses are highlighted. Copyright © 2017 John Wiley & Sons, Ltd.

Abstract Software security is an essential requirement for software systems. However, recent investigation indicates that many software development methodologies do not explicitly include methods for incorporating information security into the software development life cycles (SDLC). This research investigates, using case study, the methodologies being used in software development in Saudi Arabia and describes a model for integrating security into the SDLC. The aim is to identify the appropriate means of introducing security measures much earlier in the SDLC. This model is designed to be an extension to the existing SDLC. For achieving the research objectives and answering the research questions, the research followed a case study research design in an information‐based organization. The research identified various important elements as security standards, policies, processes being practiced, and tools used within SDLC projects. In this regard, recommendations and verification were gathered to elicit the actual activities that are appropriate to be conducted at each phase of SDLC. The non‐functional security requirements were also found, to the use of fortify and hp alm for source code review and web application testing. Copyright © 2016 John Wiley & Sons, Ltd.

The purpose of this paper is to analyze improved performance of our segmentation algorithm on IAM benchmark database in comparison to others available in the literature from accuracy and complexity points of view. Segmentation is achieved by analyzing ligatures which are strong points for segmentation of cursive handwritten words. Following preprocessing, a new heuristic technique is employed to over-segment each word at potential segmentation points. Subsequently, a simple criterion is performed to come out with fine segmentation points based on character shape analysis. Finally, the fine segmentation points are fed to train neural network for validating segment points to enhance accuracy. Based on detailed analysis and comparison, it was observed that proposed approach increased the segmentation accuracy with minimum computational complexity.