As digitization is increasing, threats to our data are also increasing at a faster pace. Generating fake videos does not require any particular type of knowledge, hardware, memory, or any computational device; however, its detection is challenging. Several methods in the past have solved the issue, but computation costs are still high and a highly efficient model has yet to be developed. Therefore, we proposed a new model architecture known as DFN (Deep Fake Network), which has the basic blocks of mobNet, a linear stack of separable convolution, max-pooling layers with Swish as an activation function, and XGBoost as a classifier to detect deepfake videos. The proposed model is more accurate compared to Xception, Efficient Net, and other state-of-the-art models. The DFN performance was tested on a DFDC (Deep Fake Detection Challenge) dataset. The proposed method achieved an accuracy of 93.28% and a precision of 91.03% with this dataset. In addition, training and validation loss was 0.14 and 0.17, respectively. Furthermore, we have taken care of all types of facial manipulations, making the model more robust, generalized, and lightweight, with the ability to detect all types of facial manipulations in videos.

This research paper introduces a novel paradigm that synergizes innovative algorithms, namely efficient data encryption, the Quondam Signature Algorithm (QSA), and federated learning, to effectively counteract random attacks targeting Internet of Things (IoT) systems. The incorporation of federated learning not only fosters continuous learning but also upholds data privacy, bolsters security measures, and provides a robust defence mechanism against evolving threats. The Quondam Signature Algorithm (QSA) emerges as a formidable solution, adept at mitigating vulnerabilities linked to man-in-the-middle attacks. Remarkably, the QSA algorithm achieves noteworthy cost savings in IoT communication by optimizing communication bit requirements. By seamlessly integrating federated learning, IoT systems attain the ability to harmoniously aggregate and analyse data from an array of devices while zealously guarding data privacy. The decentralized approach of federated learning orchestrates local machine-learning model training on individual devices, subsequently amalgamating these models into a global one. Such a mechanism not only nurtures data privacy but also empowers the system to harness diverse data sources, enhancing its analytical capabilities. A thorough comparative analysis scrutinizes varied cost-in-communication schemes, meticulously weighing both encryption and federated learning facets. The proposed approach shines by virtue of its optimization of time complexity through the synergy of offline phase computations and online phase signature generation, hinged on an elliptic curve digital signature algorithm-based online/offline scheme. In contrast, the Slow Block Move (SBM) scheme lags behind, necessitating over 25 rounds, 1500 signature generations, and an equal number of verifications. The proposed scheme, fortified by its marriage of federated learning and efficient encryption techniques, emerges as an embodiment of improved efficiency and reduced communication costs. The culmination of this research underscores the intrinsic benefits of the proposed approach: marked reduction in communication costs, elevated analytical prowess, and heightened resilience against the spectrum of attacks that IoT systems confront.

Medical images and patient information are routinely transmitted to a remote radiologist to assist in diagnosis. It is critical in e-healthcare systems to ensure that data are accurately transmitted. Medical images of a person’s body can be used against them in many ways, including by transmitting them. Copyright and intellectual property laws prohibit the unauthorized use of medical images. Digital watermarking is used to prove the authenticity of the medical images before diagnosis. In this paper, we proposed a hybrid watermarking scheme using the Slantlet transform, randomized-singular value decomposition, and optimization techniques inspired by nature (Firefly algorithm). The watermark image is encrypted using the XOR encryption technique. Extensive testing reveals that our innovative approach outperforms the existing methods based on the NC, SSIM, and PSNR. The SSIM and NC values of watermarked image and extracted watermark are close to or equal to 1 at a scaling factor of 0.06, and the PSNR of the proposed scheme lies between 58 dB and 59 dB, which shows the better performance of the scheme.

Diagnosing burns in humans has become critical, as early identification can save lives. The manual process of burn diagnosis is time-consuming and complex, even for experienced doctors. Machine learning (ML) and deep convolutional neural network (CNN) models have emerged as the standard for medical image diagnosis. The ML-based approach typically requires handcrafted features for training, which may result in suboptimal performance. Conversely, DL-based methods automatically extract features, but designing a robust model is challenging. Additionally, shallow DL methods lack long-range feature dependency, decreasing efficiency in various applications. We implemented several deep CNN models, ResNeXt, VGG16, and AlexNet, for human burn diagnosis. The results obtained from these models were found to be less reliable since shallow deep CNN models need improved attention modules to preserve the feature dependencies. Therefore, in the proposed study, the feature map is divided into several categories, and the channel dependencies between any two channel mappings within a given class are highlighted. A spatial attention map is built by considering the links between features and their locations. Our attention-based model BuRnGANeXt50 kernel and convolutional layers are also optimized for human burn diagnosis. The earlier study classified the burn based on depth of graft and non-graft. We first classified the burn based on the degree. Subsequently, it is classified into graft and non-graft. Furthermore, the proposed model performance is evaluated on Burns_BIP_US_database. The sensitivity of the BuRnGANeXt50 is 97.22% and 99.14%, respectively, for classifying burns based on degree and depth. This model may be used for quick screening of burn patients and can be executed in the cloud or on a local machine. The code of the proposed method can be accessed athttps://github.com/dhirujis02/Journal.gitfor the sake of reproducibility.

This review was undertaken to measure what effects the artificial intelligence has had on the modern-day education. AI is a field that includes machines along with the features of human intelligence features such as adaptability, learning abilities, cognitive abilities, and decision-making abilities. The result of this study is that AI is widely applied and used in various ways in education, especially in educational institutions. The inception of AI was in the form of computers which evolved into use of humanoid robots and web chatbots that individually performed the tasks and functions of instructors and used integrated computer systems along with other technologies or a trainer with them. With the help of these platforms, teachers were able to perform a host of administrative tasks like improving the quality of educational activities by conducting research and made the evaluation of student challenges more efficient and effective. Meanwhile, the system incorporates machine learning to ensure that the curriculum is tailored and personalized to an individual's need, fostering acceptance and retention, improving the learning experience, and augmenting the overall learning quality.

Wireless sensor networks (WSNs) have emerged as a promising technology in healthcare, enabling continuous patient monitoring and early disease detection. This study introduces an innovative approach to WSN data collection tailored for disease detection through signal processing in healthcare scenarios. The proposed strategy leverages the DANA (data aggregation using neighborhood analysis) algorithm and a semi-supervised clustering-based model to enhance the precision and effectiveness of data collection in healthcare WSNs. The DANA algorithm optimizes energy consumption and prolongs sensor node lifetimes by dynamically adjusting communication routes based on the network’s real-time conditions. Additionally, the semi-supervised clustering model utilizes both labeled and unlabeled data to create a more robust and adaptable clustering technique. Through extensive simulations and practical deployments, our experimental assessments demonstrate the remarkable efficacy of the proposed method and model. We conducted a comparative analysis of data collection efficiency, energy utilization, and disease detection accuracy against conventional techniques, revealing significant improvements in data quality, energy efficiency, and rapid disease diagnosis. This combined approach of the DANA algorithm and the semi-supervised clustering-based model offers healthcare WSNs a compelling solution to enhance responsiveness and reliability in disease diagnosis through signal processing. This research contributes to the advancement of healthcare monitoring systems by offering a promising avenue for early diagnosis and improved patient care, ultimately transforming the landscape of healthcare through enhanced signal processing capabilities.