Continuous remote patient monitoring is becoming critical with the increasing burden of chronic diseases, but existing wearable-based patient monitoring systems are limited in terms of accuracy and false alarms and face challenges with heterogeneous sensor data integration within the real world settings. To overcome these difficulties, an AI-based health monitoring system based on multimodal IoT and wearable signals is studied in this paper for continuous cardiac monitoring. The goal is to create a powerful and low-latency privacy-aware system that is able to reliably detect events in ambulatory environments. The methodology takes a combination of edge preprocessing to happen, uncertainty aware multimodal fusion, personalized model-adaptation and a hybrid clinician-AI labeling pipeline. Experiments using big data (600 subjects, 9 months follow up) demonstrate substantial performance improvements over 5 state of the art baselines for 0.92 sensitivity, 0.96 specificity, 0.89 PPV and 0.12 false alarms for patients per patient-day with clinically feasible latency. The results demonstrate the effectiveness of the combination of signal quality awareness fusion and adaptive personalization, which contributes to the robustness and clinical relevance in the real world. Overall, the study demonstrates evidence that AI-driven multimodal monitoring can help to move forward the standard of care for patients through proactive care and help support early detection of clinically actionable events.

This study proposed an enhanced visual explainable model for melanoma detection and risk prediction. We utilized the HAM10000 dataset, applying pre-processing techniques to improve image quality. Feature extraction and segmentation were performed using a U-Net model-based Dual Stream CNN-Transformer technique. Feature selection was optimized using the Henry Gas Solubility Optimization (HGSO) algorithm and the Water Strider Algorithm (WSA). A Deep Learning Model (DLM), specifically the Optimal Multi-Attention Fusion (MAF) ConvNeXt, was trained for melanoma detection. For disease severity prediction, we introduced a Modified ResNet-50 model combined with the Explainable AI technique Grad-CAM, providing visual explanations for the model's predictions. Experimental results demonstrate a robust classification performance with an AUC of 0.997, recall of 99%, and precision of 99.5%. This study aims to diagnose an accurate, efficient, melanoma and risk assessment. The Algorithm source code can be accessed at https://github.com/SarvachanVerma/Visual-Explanible-code-for-Melanoma_Matlab

In the published publication [...]

This Research Paper examines the implementation of blockchain-based decentralized applications (DApps) for the secure sharing of student credentials. The study begins by defining DApps and exploring how blockchain technology can enhance security and transparency in credential verification. Key aspects of blockchain technology, such as distributed ledger systems and consensus mechanisms, are discussed, along with their advantages, including immutability and decentralization. The implementation of the DApp is then reviewed, covering client-server communication interfaces and integration with existing credential systems. Additionally, the design and functionality of smart contracts are analyzed, emphasizing their role in automating processes and reducing security risks in credential sharing. Various real-world applications and use cases are presented, highlighting successful implementations. The analysis further explores the benefits and applications resulting from these implementations, as well as the challenges and limitations of adopting blockchain for credential sharing, including technical, legal, and privacy concerns. Finally, the study considers future applications of blockchain technology in education and its potential integration with emerging technologies such as artificial intelligence (AI) and the Internet of Things (IoT).

In an uncertain world filled with cyberthreats, blockchain has proven to be a revolutionary technology of significant value to most industries. While blockchain is used extensively in the fields of energy, finance and governance, healthcare is among the key sectors whose applications have been most evident as far as its adoption into these sectors is concerned. Since data is currently regarded as both an asset and currency, security has emerged as a key issue especially in healthcare as more and more data breaches have emphasized the need for enhanced planning, requirements analysis and implementation of strong cybersecurity models. This paper introduces a cloud-based blockchain architecture The approach organizes network participants into clusters, with each cluster maintaining a single copy of the blockchain. This design introduces a new blockchain architecture tailored for secure healthcare data management, significantly lowering both computational demands and communication overhead particularly when compared with conventional Bitcoin networks and existing lightweight blockchain models, while at the same time, investigating how the proposed design adequately mitigates known security threats. Experimental results show that, with increasing number of nodes, the proposed model accelerates the updates of ledger, It achieves a 63% reduction in computational load while also decreasing network traffic by 10%.

The need for surveillance has been a major concern for a long time. Maintaining the integrity and decorum of a workplace or a public area is necessary to ensure the safety and security of all the people present within its compounds. With the growing population, there is also a growing need for a better and smarter surveillance system. Unmanned Aerial Systems (UASs) are one of the most promising solutions to such a conundrum. For huge range of application authors develop an application. UASs are mobile and easily deployable. They also include features such as exact look after of the crowd behaviour by using different industrial type of sensor-based system. This article deals with the construction and deployment of surveillance device while also examining the utilization of Machine learning and drone-based technology. The main goal behind this paper is to present a UAS that can be helpful for monitoring crowd behaviour while also reducing the cost of surveillance as compared to traditional methods.

Classic Approaches of diagnosis are time consuming; A large number of farmers are unable to detect apple leaf diseases in their initial phase. This causes farmers to loss best opportunity for efficient protection and cure. Opportune and accurate Recognition of these diseases’ leftovers a notable challenge, with the principal target being the development of a strong and efficient image-based diagnostic tactic. In this study, apple leaf images were first polished using a Gaussian filter to minimize noise, followed to by segmentation to detached diseased regions. Texture features were then recovered by Gray Level Co-occurrence Matrix (GLCM) technique, which supplied as input for classifier using Decision Tree and K-Nearest Neighbors (KNN) algorithms. The outcomes Showed that the Decision Tree classifier excelled than KNN, reaching the supreme accuracy and proving its compatibility for three-class apple leaf disease classification.

Skin cancer ranks among the most concerning global health threats. Correct skin cancer diagnosis is difficult to achieve. Timely recognition of skin cancer symptoms is crucial due to the increasing incidence rates, notable mortality rates, and substantial financial burden linked with medical interventions. Assumed the severity of complications, experts have made several primary recognition methods for skin disease. Deep learning algorithms have been engineered to excel across diverse tasks and are increasingly leveraged for diagnosing skin diseases. We presents a review of deep-learning techniques for diagnosing skin disease. We initial provide a brief description of skin disease and describe some readily accessible skin datasets. Then examine well-liked deep learning architectures and well-liked frameworks aiding deep learning algorithm implementation. This article’s main goal is to offer a conceptual and comprehensive evaluation of recent efforts on deep learning-based skin disease diagnosis.

Unfathomable convolution neural networks (CNNs) have proven popular in visual super-resolution (SR). On the other hand, deep CNNs for SR usually endure from training instability, resulting in poor image SR performance. The problem can be efficiently solved by gathering additional contextual information. The authors propose a coarse-to-fine SR CNN (CFSRCNN) to recuperate high-resolution (HR) image from a low-resolution version. On benchmark datasets, extensive experiments have demonstrated that our CFSRCNN model outperforms state-of-the-art SR methods in terms of efficiency and performance.

In a piping system, pipe bends are more flexible than straight pipes because of their curved geometry, supplemented by higher stress and strain concentration, leading to one of the crucial components in piping industries. Therefore, safe design of pipe bends is essential for smooth running of the piping system, and plastic collapse moment is one of its criteria. This paper utilizes three-dimensional finite element analyses to model empirical solutions for the plastic collapse moment for different angled pipe bends subjected to combined pressure and in-plane closing, in-plane opening, and out-of-plane bending moments. Plastic collapse moments for 30 degrees to 180 degrees pipe bends are determined for elastic perfectly plastic and strain hardening materials, employing large geometry change option and internal pressure effect. It is observed from results that pressure effect is more prominent in thinner pipe bends of larger bend angle under all bending cases. For in-plane opening and out-of-plane bending moments, collapse moment increases and then decreases with increase in pressure intensity for all sizes of pipe bend. However, for in-plane opening bending moment, collapse moments keep on decreasing for thicker ( r / t = 11.33) pipe bends. Finally, the study presents new improved plastic collapse moment solutions for different angled pipe bends under bending moment and internal pressure, derived from the finite element results of elastic perfectly plastic and strain hardening material models.

Waste accumulation in the oceans poses severe environmental challenges that require efficient mechanisms for detection and removal. The methods usually employ slow and resource-intensive techniques like sonar-based detection and manual inspection, hence the need for automated mechanisms of detection. Waste detection under water is a regular monitoring task. Thus, the work proposes rapid model training which allows real-time deployment and enhancement for prevention of degradation of water bodies for various type of wastes. The work uses high-performance computing techniques such as parallel processing, NVIDIA GPUs, CUDA, and YOLO. A deep learning model featuring Convolutional Neural Networks (CNN) is integrated with YOLO (You Only Look Once) to augment detection speed along with accuracy. The model training time is particularly important for real-time applications which allows rapid iteration for the researchers and practitioners. This work points to the fact that the GPU architecture has accelerated 41X in training deep models and improved over precision, and inference time. In all aspects the proposed model claims to perform real-time waste detection within permissible computational latency. This work is in demand for autonomous systems like underwater drones which demand the fast and timely decision-making capability to operate and maneuver properly in dynamic operational environments.

This paper introduces an IIoT driven intelligent Security based robot "SENTINEL", leveraging an Arduino Uno, camera unit, ultrasonic sensor, motor controller, motors,The envisioned robot is engineered to autonomously survey a specified area, capturing images and videos with the camera unit. The ultrasonic sensor identifies obstacles to avert collisions, alerting the user. The robot maneuvers and alters its course through the motor controller and motors, managed by an Arduino Uno. This system finds utility in various domains, including surveillance and security, enhancing security-based system. It is economically developed, broadening its accessibility for future used. The system's deployment has undergone testing, demonstrating proficiency in recognizing and responding to environmental cues. A web- based interface facilitates remote control and monitoring by users.

Since the electricity system grid has developed; underground wires have been employed extensively. Because of the subsurface environment, deterioration, and rodent activity, numerous problems might arise with underground lines. Excavating the entire line is necessary to check for cable flaws, locating the source of the fault is challenging. Only that region needs to be excavated in order to find the cause of the fault, as the person who are going to repair are aware precisely which section is defective. As a result, it enables quicker maintenance of subterranean cable systems and saves a significant amount of money. Finding the underground cable faults’ distance in kilometers from the base station is the goal of this study.

As the awareness and advancement of technology is progressing and with the emergence of better capable technical concepts the easiness of day to day life is rapidly increasing. One of the important area on which human directly or indirectly depends is transportation. With the amalgamation of intelligence in the transport management system, Intelligent and smart transportation system (ITS) has evolved. The primary goal of these intelligent system is to provide traffic safety and efficient ways to reduce travelling time, travelling cost and pollution and fuel emission. This article basically focusses on the area of Internet of Vehicles (IoV) which is the latest technology in the field of intelligent transportation. Issues like internet unreliability, non-compatibility of personal devices such as cellular or tablets, very limited capability of processing along with non-presence of latest computing technologies like cloud computing. IoV is an evolving technical concept in ITS which enhances and improves the existing capabilities of existing Vehicular infrastructure. The article proposed here is focused on the domain of automated driving. The paper investigates the blind spot problem in automated driving and propose a solution for automated driving against it.

Skin-related conditions are the most common health problems worldwide. Infections carry concealed risks that may lead to a spectrum of health challenges, encompassing both physical ailments and mental health disorders such as depression. In extreme instances, it has the potential to induce skin cancer. Manually diagnosing skin conditions by medical professionals is a laborious and subjective process. Thus, it is clear to patients and physicians that automatic skin disease prediction is necessary to enable quick treatment plans. The primary aim of this study is to predict whether a skin disease is benign or malignant. To reduce image noise, our approach utilizes Gaussian filtering. Subsequently, for each image contrast correlation energy and homogeneity was computed using GLCM- Gray Level Co-occurrence Matrix. Machine learning algorithms were applied for classification: Support Vector Machine (SVM) and Decision Tree (DT). We found that Decision Tree (DT) attained a classification accuracy of 91%, which is higher related to SVM.

In today’s technological revolution, Internet of things (IoT) play very crucial role. The term IoT(Internet of Things) refers to the connection of objects to each other and to humans via the Internet. IoT is adjustable to almost any technology that is capable of providing relevant information about its operation, about the performance of an activity, and about the environmental conditions that are needed to control technology at a distance. IoT significantly improves the environment by exploding a new technology into the present one. Developed device is very easy to use and smarter than existing one. Through the use of a Wi-Fi component, the Cellphone Operated Tea Maker designed in the proposed work allows users to control their tea using their smartphones. The Tea Maker is managed by wifi methods of communication. This prototypical delivers us the feature for making tea according to our preferences and it will also be beneficial for organizations and companies by making a tea of their own choice.

Here, by employing a Barrow entropy and the standard holographic method at a cosmic framework, we formulate Barrow agegraphic dark energy (BADE), taking the Universe age as an IR cutoff scale in a flat FLRW Universe. For evaluation of statefinder parameters in [Formula: see text] and [Formula: see text] planes, trajectories have been plotted for BADE and discovered that for various values of [Formula: see text], the model exhibits both the behavior of Chaplygin gas and quintessence. Moreover, as a supplement to the statefinder study, we looked at the BADE model without interaction in the plane [Formula: see text], which might offer us a dynamic study using the energy density BADE parameter [Formula: see text] and [Formula: see text], as per VI-[Formula: see text]CDM observational data without interaction from Planck 2018 results.

The efficient selection of macro-sites for wind/solar hybrid power stations is crucial for the successful implementation of renewable energy projects. In this study, we propose the use of the MOORA (Multi-Objective Optimization on the basis of Ratio Analysis) method to facilitate the site selection process. The MOORA method allows for the simultaneous consideration of multiple criteria and objectives, making it suitable for evaluating complex decision-making problems. The site selection criteria are categorized into three main groups: environmental, technological, and geographical factors. To ensure uninterrupted power generation, wind and solar power systems are integrated into the hybrid power station design. The analysis network employs the cloud model to calculate index weights, enabling the prioritization and ranking of potential sites based on their suitability. Geographic Information System (GIS) and the Analytical Hierarchy Process (AHP) are utilized to determine the optimal locations for solar farms within the study area. The final index model incorporates a classification system that categorizes sites into "low fit," "moderate fit," "appropriate fit," and "optimal fit" based on their compatibility with the specified criteria. The proposed MOORA method and the integrated approach of GIS and AHP provide a systematic and effective framework for macro-site selection of wind/solar hybrid power stations. The results of this study contribute to the development of sustainable energy systems and support decision-making processes in the renewable energy sector.