Parenteral Drug Association

Multiple Sclerosis (MS) is an immunological disorder that causes tumors in the central nervous system. Brain Magnetic Resonance Images (MRI) were considered for the visualization of MS. In the past, neural approaches were widely utilized to detect MS, but those models failed to attain the best segmentation outcome. Moreover, traditional Deep Learning (DL) methods require massive resource data to train the system. So, a novel hybrid approach named the Buffalo-based Convolutional Neural Lesion Prediction (BCNLP) model was presented in this article to predict and segment MS lesions. In the developed model, the pre-processing function is incorporated into the hidden layer of the BCNLP. In addition, the Buffalo fitness function provides the highest prediction rate by tracking and segmenting the diseased features. Considering the other optimization model, the buffalo optimal model has future prediction behavior, and this behavior helps to track and predict the disease with high accuracy. Finally, the outcomes are determined and compared against the existing MS lesion segmentation approaches. Moreover, the improvement score in segmentation quality over the competing methods is evaluated on various metrics, such as Dice and false positive ratio. The dataset considered for validation is brain lesson MRI data from Kaggle cite. The obtained data contains 6673 images, of which 3976 are normal images, and 2697 are MS lesion images. After splitting the 70

Scene text recognition has shown considerable advancements in recent years. Despite the progress, two major challenges persist in existing techniques: false positives in text representations leading to inaccuracies in recognition, and the vast scale variability of scene texts making it difficult for networks to effectively learn from diverse samples. To address these challenges, this research proposes a modification to improve text recognition accuracy on datasets containing arbitrary and irregular text samples. The proposed approach Deep Neural Encoder-Decoder network with Probabilistic Sampling (DNED-PS), leverages various neural network including CNN (Convolutional Neural Network), to limit the extraction of irrelevant features and generate more accurate text representations. Additionally, the integration of a Deep Neural Encoder-Decoder network with a modified transformer facilitates accurate text sequence generation in a bidirectional workflow, further improving overall text recognition performance. Further, the proposed model consists of three sections: Adaptive Convolutional Neural Network (CNN), Localization and Resampling (LNR), and an Encoder–Decoder network with Probabilistic Sampling (DNED-PS). Experimental evaluations demonstrate that DNED-PS achieves state-of-the-art performance across three challenging irregular text datasets. On the IIIT5K dataset, DNED-PS attains an accuracy of 98.3

With the widespread use of cloud computing (CC), data security has arisen as a key concern owing to risks connected to centralized storage and third-party data management. Traditional encryption solutions, while successful in some cases, frequently confront scalability, key management, and real-time access control (AC) issues, particularly in remote cloud systems. This research presents a better Blockchain-Integrated Optimized Cryptographic Framework (BIOCF), which aims at data privacy through Paillier Homomorphic Encryption (PHE) and entails a hybrid optimization model-Greylag Goose Optimization (GGO) and Crayfish Optimization (CO)-for dynamic cryptographic key generation and management. The integration of blockchain technologies for decentralized key management, thus allowing for threshold cryptography for key-sharing mechanisms and smart contracts for AC, significantly enhances security and reduces the reliance on centralized authority. Moreover, the cryptographically hashed encrypted data stored on the blockchain offers a strong mechanism for checking data integrity. The developed model is experimentally validated using existing cryptographic model techniques. In BIOCF, the proposed model achieved an extremely low encryption time of about 1.23 s and a decryption time within 1.6 s for data on up to 10 MB. For that, the key generation is recorded in an efficient manner of 0.684 s, along with an overhead of 2.5 s. The resource utilization during the key generation is to be 48.5 %. Highest throughput of up to 140.23 kb/s can also be demonstrated at 250 TPS. The model achieved an MAPE of 4.85 % for encryption time, with an MAE of 0.12, an RMSE of 0.16, and an R2 score of 0.842.

With the growth in the number of vehicular applications, there has been significant need for more processing and communication capabilities in the heterogeneous systems. Vehicular Ad-Hoc Networks or simply VANETs have been established with the aim of improving traffic congestion, accidental rates, and safety for drivers through vehicle-to-vehicle (V2V) communication and vehicle-to-infrastructure V2I communication. However, as a result of constant changes in the vehicular environment due to vehicle dynamics to include mobility, limited broadcast range and the adaptive network formation, there are sets back in scalability and performance. Another challenge that affects the adoption of autonomous driving systems and the Internet of Vehicles (IoV) is security threats, privacy issue, and data management challenges. This paper reviews the use of fog computing and artificial intelligence (AI) techniques to solve these challenges. Fog computing provides the efficient and distributed processing capacity and ML facilitates intelligent decision-making in real-time applications such as traffic flow analysis, obstacle identification, and prediction of the safety measures required in a vehicle. Thus, the paper discusses how such technologies can help in increasing traffic safety in IoV, coordinating the real-time data analysis, and implementing efficient resource utilization. In addition, the future innovations like blockchain mobility and artificial intelligence are included and their use in mitigating the problems in existing vehicular communication networks noted. In this paper, the author’s objective is to review the advancements in technology in relation to smart transport systems and the possibility of their development in the future.

The pursuit of sustainable and clean energy sources has prompted significant research efforts toward developing efficient photocatalytic materials for hydrogen production. In this study, we present a comprehensive review of the synthesis of BNiO3 nanocomposites and their potential application as efficient photocatalysts for hydrogen production. The synthesis of BNiO3 nanocomposites involves the integration of bismuth oxide (Bi2O3) and nickel oxide nanoparticles with boron nitride nanosheets. Various synthesis techniques have been employed to fabricate these nanocomposites, including sol–gel, hydrothermal, and co-precipitation methods. The choice of synthesis method significantly influences the nanocomposites' structural, morphological, and optical properties, thereby affecting their photocatalytic performance. The morphological characterization techniques, such as scanning electron microscopy, transmission electron microscopy, and X-ray diffraction, have been utilized to investigate the structural and morphological properties of BNiO3 nanocomposites. The photocatalytic activity of BNiO3 nanocomposites for hydrogen production has been extensively studied. The mechanism of hydrogen production involves the absorption of solar energy by the BNiO3 nanocomposites, followed by the generation of electron–hole pairs. This report provides valuable insights into the synthesis techniques, characterization methods, and photocatalytic performance of BNiO3 nanocomposites. Further research is warranted to optimize the synthesis parameters and explore novel strategies for enhancing the efficiency and stability of these nanocomposites, ultimately contributing to the development of sustainable energy solutions.