The fast growth of social media platforms has resulted in an unprecedented increase in multimodal material, which includes text, images, and audio. Traditional detection methods, which are sometimes limited to single modalities, fail to capture the intricate character of hate speech, in which tone, visual clues, and contextual subtleties all play important roles. In this study, we present a complete multimodal hate speech detection method that combines text analysis, image identification, and audio feature extraction. Our technique, which draws on advances in machine learning and natural language processing, integrates various modalities to improve detection accuracy and provide a more comprehensive understanding of the situation. We discuss the limits of imbalanced datasets, as well as the complicated, scattered nature of hate speech. Our proposed model uses cross- modal learning to align features and bridge semantic gaps between modalities, resulting in a robust solution for recognizing hostile material with greater precision and reliability. This work lays the groundwork for scalable, multimodal moderation technologies capable of dealing with the complex nature of online hate speech.

Chronic Kidney Disease (CKD) is a critical global health concern, especially in low- and middle-income nations, where compromised kidney function hinders effective blood filtration. Early detection is imperative as 10% of the world's population suffers from CKD, leading to increased fatalities due to limited access to affordable treatment. Over the past two decades, CKD- related deaths surged by 82.3 %, highlighting its significant contribution to global mortality, as per the 2010 world-wide Burden of Disease Study. The challenge is compounded by CKD's often asymptomatic early stages, emphasizing the need for routine testing as a vital diagnostic tool. Early identification is crucial in preventing End-Stage Renal Disease (ESRD), saving lives, and facilitating efficient treatment. This study focuses on developing a CKD predictive model using machine learning methods, analyzing a dataset with 9993 rows and 25 columns containing key kidney health information. Various methods, including Random Forest, Logistic Regression, Decision Tree, Support Vector Machine (SVM), k-Nearest Neighbours (KNN), and Naive Bayes, are compared for optimal CKD prediction. Addressing missing data ensures result accuracy. The ultimate goal is to provide a reliable, cost-effective model for early CKD identification, aiding medical professionals and patients by expediting diagnosis and enabling timely intervention.

OBJECTIVES:Vitamin D is essential for maintaining bone and mineral balance. This study aims to identify the most effective route for achieving optimal vitamin D levels (≥30 ng/mL) to support bone and mineral health. METHODS:In this open-label randomized trial, 132 participants aged 18-60 with initial serum vitamin D levels below 30 ng/mL were divided into three intervention groups: daily 800 I.U. oral tablet (Group A), weekly 60,000 I.U. oral sachet (Group B), and monthly 300,000 I.U. intramuscular injection (Group C). The study assessed changes in their serum vitamin D levels at six and 12 weeks. RESULTS:The monthly intramuscular (IM) group consistently had the highest mean vitamin D levels at six weeks 38.38(±9.953) (p<0.002) and 12 weeks 48.15(±7.71) (p<0.001). Vitamin D insufficiency was reduced to 34.8 % at six weeks (p=0.434) and 6.8 % at 12 weeks (p=0.002). Notably, 100 % of the monthly IM group achieved vitamin D sufficiency at 12 weeks. CONCLUSIONS:The monthly IM route demonstrated superior effectiveness compared to tablets and sachets at both the 6-week and 12-week points. A significantly larger number of monthly IM participants achieved vitamin D sufficiency compared to the other groups.

There is a recent trend that makes Node.js a necessary part of a JavaScript client-side application and configuring a Node.js server is still a complicated and time-consuming task, especially for beginner web developers. This work presents an abstract syntax tree-based cloud computing solution to simplify the setup of a Node.js server. Our software reduces the overall configuration time and incorporates an AST-based program creation algorithm to automate the process. Additionally, our software supports input in YAML or JSON formats. We analyze how cloud computing increases productivity and accessibility, enabling full-smooth distributed web development. We also evaluate the benefits of AST-based code generation, including enhanced performability, manageability, and maintainability with a focus on popular use cases and benchmarks. Our product’s scalability and flexibility is enhanced since it is relocated to the cloud.

Introduction:Remdesivir is currently approved for treating hospitalised patients with COVID-19. However, it is a priority to monitor its safety and effectiveness in various clinical settings. This study was undertaken to assess the impact of remdesivir on inflammatory and prognostic markers of COVID-19.Materials and Methods:A hospital-based prospective longitudinal study was conducted over two months comprising event monitoring of COVID-19 patients administered remdesivir as per standard guidelines. The demographic details, risk factors and all baseline parameters were collected. The patients were followed up for the appearance of any adverse drug reactions (ADRs) after the start of remdesivir therapy from Day 1 to discharge or death every day. Repeat Lab tests were done on days 2, 4, 6 and 10 days to assess the impact of remdesivir on inflammatory and prognostic markers of COVID-19 over time. Significant predictors of survival in the cohort were also assessed.Results:A total of 60 COVID-19 patients were administered remdesivir. The mean age of the patients was 59.2 (+13.7) years. There was a significant improvement in the serum creatinine (decreased from 0.9 to 0.7 mg/dL), lymphocyte count {decreased from 9.2 to 7.3 (109 cells/L)} and serum sodium (increased from 134.6 to 137.4) of the patients over six days after the administration of remdesivir. The significant survival predictors were multiple organ failure (P 0.046) and WBC count on Day 10 (P 0.001).Conclusion:Remdesivir administration improved the prognostic biomarker profile in COVID-19 patients.

Recognizing facial expressions is a major challenge and will be required in the latest fields of research such as the industrial Internet of Things. Currently, the available methods are useful for detecting singular facial images, but they are very hard to extract. The main aim of face detection is to capture an image in real-time and search for the image in the available dataset. So, by using this biometric feature, one can recognize and verify the person’s image by their facial features. Many researchers have used Principal Component Analysis (PCA), Support Vector Machine (SVM), a combination of PCA and SVM, PCA with an Artificial Neural Network, and even the traditional PCA-SVM to improve face recognition. PCA-SVM is better than PCA-ANN as PCA-ANN has the limitation of a small dataset. As far as classification and generalization are concerned, SVM requires fewer parameters and generates less generalization errors than an ANN. In this paper, we propose a new framework, called FRS-DCT-SVM, that uses GA-RBF for face detection and optimization and the discrete cosine transform (DCT) to extract features. FRS-DCT-SVM using GA-RBF gives better results in terms of clustering time. The average accuracy received by FRS-DCT-SVM using GA-RBF is 98.346, which is better than that of PCA-SVM and SVM-DCT (86.668 and 96.098, respectively). In addition, a comparison is made based on the training, testing, and classification times.

Many real life problems that play a major role in human lives are mostly optimization problems and need to be solved in order to judiciously utilize vital resources. The vast solution space consists of a large number of feasible solutions. Solving these problems requires finding the most optimal solution while satisfying the constraints imposed (if any). Vehicle Routing is a real life problem originated primarily in the logistics industry where the consignments are to be delivered to the clients in such a way, that there is minimal usage of resources like fuel, and time. In addition, the consignments are to be successfully delivered to the clients through the shortest route and utilizing the maximum capacity of each vehicle. Automation is the key to solve such large problems while reducing effort and complexity of the solution. Literature reveals that nature inspired algorithms have proved their ability for solving such large complex optimization problems. These algorithms are inspired from various natural phenomena and are supported by their successful survival. This paper proposes a hybrid framework to solve Vehicle Routing Problem (VRP) utilizing Differential Evolution and Marine Predators algorithm. We have considered the variant called Capacitated Vehicle Routing Problem (CVRP) to conduct experiments and compared the results to evaluate the performance of proposed hybrid approach DEMPA with Differential Evolution (DE). Results indicate the superiority of the proposed approach over Differential Evolution.

Healthcare sectors are gradually accepting technology that provides various remote healthcare facility, disease prediction, and in-home diagnostics capabilities, which combine Machine Learning (ML) and the Internet of Things (IoT). ML offers tools for management of electronic records, integration of data, and techniques for computer-aided diagnosis which helps in disease diagnosis, prediction, and treatment suggestion. The IoT connect all type medical devices, patients monitoring tools, and wearable devices that can send immediate data to concern authority like doctor. In this chapter, authors focus on ML applications for IoT-based healthcare. In this chapter, first, we introduce the basic concepts of ML and IoT and summarize the advantages of the techniques over traditional approaches of healthcare. We described the challenges of using ML and IoT in research and finally presented applications of ML like identifying diseases and diagnosis, drug discovery, personalized medicine, and smart health records that are augmented with IoT that significantly have an impact on delivery of healthcare.

Cloud computing has brought a new revolution to IT industry. In this era, information sharing is more convenient and fast, but applying security mechanism on information while outsourcing it on cloud is challenging. In this paper, we proposed hybrid approach comprising of symmetric and asymmetric algorithm. For encrypting file, we generated optimal and efficient symmetric key using genetic algorithm (GA), further an efficient and secure re-encryption concept has been proposed for ensuring the security of data in unreliable cloud environment. Our results have proved that our proposed approach is more efficient and secured while sharing file with authorised users using cloud in comparison to existing approaches.

Biomedical image classification is an important task to generate classification maps for deadly diseases like cancer, tumor, etc. In this paper, a novel technique for classification of biomedical images is proposed. The proposed technique takes raw biomedical image as input and applies indexing to retrieve the parameters used to classify different diseases based on deep learning. The proposed Indexing-Based Peculiarity Extrication from biomedical images using deep learning (IBPE) is evaluated using GLCM, Gabor filter, fuzzy clustering means and SVM. To achieve an effective biomedical image classification, this framework system isolates its work in various stages; these phases are important to give the better classification accuracy. Evaluation proves the efficiency of the proposed IBPE with an achieved accuracy of 96.7 over 92.5% of the existing method.

Ovarian ectopic is a rare form of ectopic pregnancy and reaching up to term is an extreme rarity. It is usually diagnosed incidentally when a cesarean section is done for other obstetric causes. Being highly vascular, it may result in obstetric catastrophes. We report first live post-term ovarian pregnancy well managed with multidisciplinary approach. A 35-year multigravida at 44 weeks presented with a diagnosis of placenta percreta. Magnetic resonance imaging suggested abdominal pregnancy with a live fetus. On laparotomy, it was live ovarian ectopic with placenta attached to left infundibulopelvic vessels and later confirmed on histological examination. Ovarian ectopic is best diagnosed in the first trimester by ultrasound and managed by laparoscopy or laparotomy. It is usually misdiagnosed as an abdominal pregnancy at term even with the use of high-end technology. It always poses a dilemma for clinicians. Preoperative magnetic resonance imaging helps in the planning of surgery with all precautionary measures and counseling of patients.

Cloud computing is an emergent computing paradigm; however, data security is a significant issue in recent time while outsourcing the data to the cloud preventing users to upload their data on cloud. The data forwarded to cloud can be protected using some cryptographic techniques based on identity, attributes, and prediction. But these algorithms lack their performance when a revoked user collude with cloud; therefore, it becomes essential to develop a secure data sharing framework with some enhanced cryptographic techniques. The proposed methodology presented a secure privacy preserving data sharing with encryption technique called dynamic unidirectional proxy re-encryption (PRE) with cipher text policy attribute-based encryption. The technique ensures the privacy, integrity, and security of the data while retrieving through the cloud. The framework is implemented in the cloud sim with java language. Experimental results proved that proposed frame work attains reasonable results compared to traditional methods.

Stock market or Share market is one of the most complicated and sophisticated way to do business. Small ownerships, brokerage corporations, banking sector, all depend on this very body to make revenue and divide risks; a very complicated model. However, this paper proposes to use machine learning algorithm to predict the future stock price for exchange by using open source libraries and preexisting algorithms to help make this unpredictable format of business a little more predictable. We shall see how this simple implementation will bring acceptable results. The outcome is completely based on numbers and assumes a lot of axioms that may or may not follow in the real world so as the time of prediction.

The present study was aimed to characterize linseed accessions for genetic diversity, population structure and establish marker-trait association using SNP markers. A total of 10057 SNPs were identified in 86 accessions through genotyping by sequencing approach of next-generation sequencing. The polymorphism information content (PIC) value was found to be in the range from 0.08 to 0.30 with an average of 0.27 ± 0.09. Moderate value of PIC (0.30–0.60) was shown by 48% SNPs while remaining had low PIC values. Based on Jaccard’s similarity coefficient, the genetic distance varied from 0.17 to 0.65 with an average of 0.42 ± 0.07. Based on genetic relatedness among the accessions, CG 79, CG 86, GP 25, GP 31 GP 33, GP 47, and GP 49 were found to be most divergent and could be potential accessions for the future breeding program of linseed to create further genetic variability. The neighbor-joining clustering revealed that 86 genotypes were grouped into four clusters with 28, 8, 21 and 29 accessions respectively. An admixture model-based clustering method in STRUCTURE was also implemented which grouped all the accessions in four subpopulations (K = 4) as similar to NJ clustering. One pleiotropic SNPs was observed for capsule weight/plant and seed weight/plant which could assist in their simultaneous improvement during the breeding programme.

The Cloud computing develops as another processing worldview which intends to give dependable, tweaked and QoS ensured dynamic registering conditions for end-clients. In this paper, we ponder the Cloud computing worldview from different angles, for example, definitions, unmistakable highlights, and empowering advancements. This paper brings a presentation survey on the Cloud computing and give the cutting edge of Cloud computing advancements. Keywords: Cloud Computing, Grid Computing, Cyberinfrastructure, Distributed Computing

Cloud computing is the emergent technology that face one of the significant issues time with data security while outsourcing the data onto the cloud in recent. Some cryptographic techniques have been used for protection in form of identity, attributes and prediction algorithms nonetheless these algorithms lack their performance and becomes are very prone to attackers when an unauthorized user reunited the system with dissimilar way for privileges to the similar data files. The essential need of this data security solved by some enhanced cryptographic techniques in DRM utilizing a secure privacy preserving data sharing with encryption techniques of Dynamic Unidirectional Proxy Re-Encryption. This technique is based on Cipher text Policy Attribute by providing the privacy, integrity and security of the data while retrieving.