No abstract

No abstract

In this paper, a novel machine learning model is proposed to predict the staying time of international migrants. The competitive machine learning approaches which can be used to predict the staying time of international migrants suffer from hyper-attributes tuning and over-fitting issues. Therefore, a particle swarm optimization (PSO) based support vector machine (SVM) model is proposed to predict the staying time of international migrants. Extensive experiments are performed by considering the international migrants dataset to predict the staying time of international migrants. Experimental results illustrate that the proposed approach outperforms the existing machine learning approaches in terms of f-measure, accuracy, specificity, and sensitivity.

No abstract

Introduction Drug repurposing is the need of the hour considering the medical emergency caused by the COVID-19 pandemic. Recently, cytokine storm by the host immune system has been linked with high viral load, loss of lung function, acute respiratory distress syndrome (ARDS), multiple organ failure, and subsequent fatal outcome. Objective This study aimed to identify potential FDA approved drugs that can be repurposed for COVID-19 treatment using an in-silico analysis. Methods In this study, virtual screening of selected FDA approved drugs was performed by targeting the main protease (M pro ) of SARS-CoV-2 and the key molecules involved in the ‘Cytokine storm’ in COVID-19 patients. Based on our preliminary screening supported by extensive literature search, we selected FDA approved drugs to target the SARS-CoV-2 main protease (M pro ) and the key players of cytokine storm, TNF-α, IL-6, and IL-1β. These compounds were examined based on systematic docking studies and further validated using a combination of molecular dynamics simulations and molecular mechanic/generalized/Born/Poisson-Boltzmann surface area (MM/G/P/BSA) free energy calculations. Results Based on the findings, Rifampicin and Letermovir appeared as the most promising drug showing a very good binding affinity with the main protease of SARS-CoV-2 and TNF-α, IL-6, and IL-1β. However, it is pertinent to mention here that our findings need further validation by in vitro analysis and clinical trials. Conclusion This study provides an insight into the drug repurposing approach in which several FDA approved drugs were examined to inhibit COVID-19 infection by targeting the main protease of SARS-COV-2 and the cytokine storm. Graphic abstract Supplementary Information The online version contains supplementary material available at 10.1007/s43440-021-00228-0.

COVID-19 pandemic has now expanded over 213 nations across the world. To date, there is no specific medication available for SARS CoV-2 infection. The main protease (Mpro) of SARS CoV-2 plays a crucial role in viral replication and transcription and thereby considered as an attractive drug target for the inhibition of COVID-19,. Natural compounds are widely recognised as valuabe source of antiviral drugs due to their structural diversity and safety. In the current study, we have screened twenty natural compounds having antiviral properties to discover the potential inhibitor molecules against Mpro of COVID-19. Systematic molecular docking analysis was conducted using AuroDock 4.2 to determine the binding affinities and interactions between natural compounds and the Mpro. Out of twenty molecules, four natural metabolites namely, amentoflavone, guggulsterone, puerarin, and piperine were found to have strong interaction with Mpro of COVID-19 based on the docking analysis. These selected natural compounds were further validated using combination of molecular dynamic simulations and molecular mechanic/generalized/Born/Poisson-Boltzmann surface area (MM/G/P/BSA) free energy calculations. During MD simulations, all four natural compounds bound to Mpro on 50ns and MM/G/P/BSA free energy calculations showed that all four shortlisted ligands have stable and favourable energies causing strong binding with binding site of Mpro protein. These four natural compounds have passed the Absorption, Distribution, Metabolism, and Excretion (ADME) property as well as Lipinski’s rule of five. Our promising findings based on in-silico studies warrant further clinical trials in order to use these natural compounds as potential inhibitors of Mpro protein of COVID.

Content-based image retrieval (CBIR) system generally retrieves images based on the matching of the query image from all the images of the database. This exhaustive matching and searching slow down the image retrieval process. In this paper, a fast and effective CBIR system is proposed which uses supervised learning-based image management and retrieval techniques. It utilizes machine learning approaches as a prior step for speeding up image retrieval in the large database. For the implementation of this, first, we extract statistical moments and the orthogonal-combination of local binary patterns (OC-LBP)-based computationally light weighted color and texture features. Further, using some ground truth annotation of images, we have trained the multi-class support vector machine (SVM) classifier. This classifier works as a manager and categorizes the remaining images into different libraries. However, at the query time, the same features are extracted and fed to the SVM classifier. SVM detects the class of query and searching is narrowed down to the corresponding library. This supervised model with weighted Euclidean Distance (ED) filters out maximum irrelevant images and speeds up the searching time. This work is evaluated and compared with the conventional model of the CBIR system on two benchmark databases, and it is found that the proposed work is significantly encouraging in terms of retrieval accuracy and response time for the same set of used features.

Everyone’s life on earth influenced by a global coronavirus outbreak COVID- 19. Two regular practices, pathology tests, and Computer Tomography (CT) scan used to diagnose COVID-19. Pathology tests produce a considerable amount of false-positives & are time-consuming, whereas CT scans tests are costly and require expert advice. Hence, the main aim of this work is to develop a fast, accurate, and low-cost diagnostic system for detection of COVID-19 using inexpensive chest X-rays and the modern Deep Convolutional Neural Network(CNN) approach to assist medical professionals. In this study, two pre-trained CNN models (VGG16 and InceptionV3) are evaluated by several experiments using data augmentations. The analysis is based on 2905 images of chest X-rays with 219 confirmed positive COVID-19 and 1345 positive pneumonia cases taken from the open-source database consisting of patients suffering from the COVID-19 disease. Since a database consists of multiple types of diseases, multiclass classification for diagnosis of COVID-19 is used. The InceptionV3 model provides the highest classification accuracy (99.35% and 98.29%) for two binary classifications (normal vs. COVID-19 and COVID- 19 vs. Pneumonia) compare to VGG16 model’s accuracy (97.71% and 96.27%). Whereas, VGG16 provides highest accuracy (98.84%)for multiclass-classification(normal vs COVID- 19 vs pneumonia) as compared to VGG16 model’s accuracy(96.35%).