ABSTRACT
Background: At the onset of a pandemic, such as COVID-19, data with proper labeling/attributes corresponding to the new disease might be unavailable or sparse. Machine Learning (ML) models trained with the available data, which is limited in quantity and poor in diversity, will often be biased and inaccurate. At the same time, ML algorithms designed to fight pandemics must have good performance and be developed in a time-sensitive manner. To tackle the challenges of limited data, and label scarcity in the available data, we propose generating conditional synthetic data, to be used alongside real data for developing robust ML models. Methods: We present a hybrid model consisting of a conditional generative flow and a classifier for conditional synthetic data generation. The classifier decouples the feature representation for the condition, which is fed to the flow to extract the local noise. We generate synthetic data by manipulating the local noise with fixed conditional feature representation. We also propose a semi-supervised approach to generate synthetic samples in the absence of labels for a majority of the available data. Results: We performed conditional synthetic generation for chest computed tomography (CT) scans corresponding to normal, COVID-19, and pneumonia afflicted patients. We show that our method significantly outperforms existing models both on qualitative and quantitative performance, and our semi-supervised approach can efficiently synthesize conditional samples under label scarcity. As an example of downstream use of synthetic data, we show improvement in COVID-19 detection from CT scans with conditional synthetic data augmentation.
ABSTRACT
Coronavirus disease 2019 (COVID-19) has so far been the most devastating pandemic ever faced by mankind. Caused by the highly transmissible severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), the disease is becoming deadly due to frequent emergence of variants. The virus belongs to the group, Betacoronaviruses, and shares more than 90% amino acid identity with SARS-CoV. The SARSCoV-2 possess a single-stranded positive-sense RNA which is the largest known viral RNA genome consisting of 25,000-30,000 nucleotides with 14 ORFs. The 3'-region of the genome harbours four structural proteins, namely;the spike, nucleocapsid, envelope and the membrane proteins;the S protein plays the most important role during infection. Genomics-led studies are pre-requisites to understand the pathogenicity of any pathogen and for devising its management strategies. The availability of SARS-CoV-2 sequence data and suitable bioinformatics platforms have allowed researchers to identify potential therapeutic targets and to predict immune response for accelerating therapeutics and vaccine development. A plethora of such options are available that includes repurposing existing drugs, monoclonal antibodies, anti-inflammatory agents, etc. Moreover, different types of vaccines such as mRNA-based, viral vector, inactivated virus, etc., with different efficacy levels have been approved. However, their efficacy might get compromised with time, particularly due to frequent mutations in the viral genomes. Here, we provide a comprehensive insight into the genome structure, evolution, pathogenicity as well as the achieved success and limitations in management of this notorious virus.
ABSTRACT
The demand for a smart classroom has been compounded by Covid-19, which allows students to have a meaningful learning experience while staying home. Students who join a classroom in online mode don't have the opportunity to experience a classroom setting because of the hybrid mode of teaching (both online and offline classes). As a result, they have problems such as not being able to see the board clearly, not being able to follow the lecturer because he or she is out of frame, and thus having difficulty learning. Furthermore, this results in lower interaction between the online students and the professor. To teach effectively, the professor is unable to use the entire length of the board as it would not be visible to students joining in online mode. As students and instructors, we identified the issue and developed a plug-and-play device which is portable to address the aforesaid problem during this testing and difficult period of time of the pandemic. The paper outlines the practical implementation of a plug-and-play device that meets the aforementioned requirements. The model also considers power usage, as it can dynamically control energy-consuming resources such as lighting and air conditioning in response to the environment and the presence of students. © 2022 IEEE.
ABSTRACT
Purpose The present study is mainly concerned with investigating the migration to online learning under the coronavirus disease 2019 (COVID-19) pandemic and analysing the adoption of technology in the context of Indian educational organisations. The purpose of the paper is to identify aspects that explain and predict the adoption propensity of new technology by users as a dependent variable, with perceived usefulness (PU) and perceived ease of use (PE) as independent variables and personality and self-efficacy as the moderator variables. Design/methodology/approach An online as well as offline survey is collected from N = 202 employees (teachers/faculty) from private (N = 97) and public (N = 105) educational organisations located in India. A conceptual model of technology adoption is developed and validated, measuring the impact of Big Five personality factors and self-efficacy on technology adoption. Findings Results of moderation analysis suggest that personality traits moderate the relationship between PU, PE and acceptance of technology (TAP). Originality/value The present research uniquely contributes to the limited literature on the role of personality and self-efficacy in adopting technology and the outcomes. Furthermore, the research captures the theoretical and practical understanding of the PU, PE and TAP link in educational organisation and COVID-19 context.
ABSTRACT
The whole world at present is under the grasp of a pandemic termed as COVID-19. The World Health Organization (WHO) guidelines suggest that the social distancing norms are followed with contactless operations as far as possible. Therefore, the population around the world is turning towards efficient modes of operating the daily work with minimal human contact. To contain the spread of the novel coronavirus or COVID-19, it is important and suitable to deploy machinery for operating in conditions wherever social distancing is required. The multipurpose robot makes it feasible to minimize human contact and carry out operations without the risk of the spread of the virus. This chapter aims at the fabrication of a robot that can have multiple utilities and is employed in different areas as per the requirement of the user. © 2021 Scrivener Publishing LLC.