Generative Data Augmentation for Non-IID Problem in Decentralized Clinical Machine Learning

Swarm learning (SL) is an emerging promising decentralized machine learning paradigm and has achieved high performance in clinical applications. SL solves the problem of a central structure in federated learning by combining edge computing and blockchain-based peer-to-peer network. While there are promising results in the assumption of the independent and identically distributed (IID) data across participants, SL suffers from performance degradation as the degree of the non-IID data increases. To address this problem, we propose a generative augmentation framework in swarm learning called SL-GAN, which augments the non-IID data by generating the synthetic data from participants. SL-GAN trains generators and discriminators locally, and periodically aggregation via a randomly elected coordinator in SL network. Under the standard assumptions, we theoretically prove the convergence of SL-GAN using stochastic approximations. Experimental results demonstrate that SL-GAN outperforms state-of-art methods on three real world clinical datasets including Tuberculosis, Leukemia, COVID-19. © 2022 IEEE.

Towards Privacy-Preserving Computation on Gene: Construct Covid-19 Phylogenetic Tree using Homomorphic Encryption

Constructing a phylogenetic tree is an essential method of analyzing the evolution of the covid-19 virus. In the case of multiple entities holding different coronavirus genetic data, it is simple to aggregate all data into one entity and then calculate the phylogenetic tree. However, such a method is challenging to carry out. Genetic data is susceptible and has high economic value, and it is usually impossible to copy between different entities directly. Also, the direct sharing of genetic data can lead to data leaks or even legal problems. In this paper, we propose a homomorphic-encryption-based solution to tackle this problem, where two participants, A and B, both hold a part of covid-19 genetic data and compute the gene distance matrix calculation of the overall dataset without revealing the genetic data held by both parties. After the computation, participant A can decrypt the final distance matrix from the encrypted result and then use the plain-text result to construct the covid-19 phylogenetic tree. Experiment results show that the proposed method can process the genetic data accurately in a short time, and the phylogenetic tree generated by the proposed solution has no loss of accuracy compared to plain-text calculation. In terms of engineering optimization, we propose an optimized encryption method, which can further shorten the encryption time of the entire dataset without reducing the security level. © 2022 IEEE.

SARS-CoV-2 membrane protein is incompetent at inducing antibody-dependent enhancement though Fc receptor pathway

Objective To investigate the immune characteristics of SARS-CoV-2 membrane (M) protein, especially the possibility of inducing antibody-dependent enhancement effect (ADE) .Methods Full-length SARS-CoV-2 M protein was prepared by prokaryotic expression system and purified.BALB/ c mice were immunized subcutaneously three times (on day 1, day 14 and day 21) by purified M protein.Serum samples were collected before immunization and after each immunization.The specificity of immune sera against M protein was identified by Western blot, and the antibody titers were detected by ELISA and neutralization test.In the presence of anti-M protein serum, the proliferation of SARS-CoV-2 in dendritic cells, nature killer cells, T and B cells was detected in vitro.Results The immune sera from BALB/ c mice immunized with purified full-length M protein of SARS-CoV-2 specifically recognized viral M protein.The titer of anti-whole virus antibody in immune sera was about 1 ∶ 400, but the antibody could not neutralize live virus.Moreover, the antibody could not help the virus to infect and proliferate in the various types of immune cells with Fc receptor (FcR).Conclusions Non-neutralizing antibody induced by M protein could not cause ADE through FcR pathway.

Exploring privacy paradox in contact tracing apps adoption

Purpose: Understanding the privacy concerns of individuals in the adoption of contact tracing apps is critical for the successful control of pandemics like COVID-19. This paper explores the privacy paradox in the adoption of contact tracing apps in Australia. Design/methodology/approach: A comprehensive review of the related literature has been conducted, leading to the development of a conceptual model based on the privacy calculus theory and the antecedent-privacy concern-outcome framework. Such a model is then tested and validated using structural equation modelling on the survey data collected in Australia. Findings: The study shows that perceived benefit, perceived privacy risk and trust have significant influences on the adoption of contact tracing apps. It reveals that personal innovativeness and trust have significant and negative influences on perceived privacy risk. The study further finds out that personal innovativeness is insignificant to perceived benefit. It states that perceived ease of use has an insignificant influence on perceived privacy risk in the adoption of contact tracing apps. Originality/value: This study is the first attempt to use the privacy calculus theory and the antecedent–privacy concern–outcome framework for exploring the privacy paradox in adopting contact tracing apps. This leads to a better understanding of the privacy concerns of individuals in the adoption of contact tracing apps. Such an understanding can help formulate targeted strategies and policies for promoting the adoption of contact tracing apps and inform future epidemic control through effective contact tracing for better emergency management. © 2022, Emerald Publishing Limited.

Riding on the waves of the COVID-19 pandemic in re-thinking organizational design: a contingency-based approach

Purpose: This paper aims to explore what organizational structural designs and strategies that organizations can seek to adopt so as to enable them to respond effectively to the post-COVID-19 environment conditions. It adopts the contingency theory, which asserts that organizational survival is dependent on the fit between organizational structures and contingencies. Furthermore, the paper applies Miles et al. (1978) typology of business strategy to study four strategic orientations that organizations can adopt in achieving better organizational performances. Design/methodology/approach: A framework of six strategic orientation archetypes is proposed that can support organizations in re-thinking their organizational structural designs for building up and strengthening resilience during the COVID-19 pandemic. The authors explore the influence of transactional leadership and transformational leadership and organizational culture on the adoption of strategic orientation. In addition, the authors developed six propositions. Findings: Organizations that have a prospector orientation tend to focus on creativity and innovation. Organizations that have a defender orientation tend to focus on reducing manufacturing and distribution costs and maintaining or improving product quality. Analyzers tend to be second-movers after prospectors making slower and fewer changes to their products. Originality/value: To the authors’ best understanding, this study is one of the first to explore the interrelationship between organizational structures, situational factors and strategic orientation. © 2022, Emerald Publishing Limited.

[Analysis of epidemiological characteristics of human coronavirus in hospitalized children with respiratory tract infection in Hebei region].

Objective: To investigate the epidemiological characteristics of human coronavirus (HCoV) in hospitalized children with respiratory tract infection in Hebei region, providing evidence for the diagnosis and prevention of children with respiratory tract infection. Methods: A retrospective study was conducted on 1 062 HCoV positive children hospitalized for respiratory tract infection in Children's Hospital of Hebei Province from January 2015 to December 2020, aged from 33 days to 14 years, with a median age of 2 years. 27 932 (60.9%) were males and 17 944(39.1%) were females. And the gender, ages, seasonal distribution, HCoV-positive rates, co-detection distribution and clinical diagnosis of HCoV positive cases were analyzed by SPSS 25.0. Enumeration data were expressed by frequency and percentage; categorical variable were compared by the Pearson χ2test. Results: The overall HCoV-positive rate was 2.31% (1 062/45 876), which was 2.37% (662/27 932) in male children and 2.23% (400/17 944) in female children. There was no statistically significant difference between genders (χ²=0.916, P=0.339). Children at age groups<1 years (2.44%) and 1-<3 years (2.63%) had higher HCoV-positive rates than those at age groups 3-<5 years (1.97%) and ≥5 years (1.38%) (χ²=27.332,P<0.01). The HCoV-positive rates from 2015 to 2018 were 2.13%, 2.45%, 2.28% and 2.23%. The HCoV-positive rate of 2019 (1.71%) was significantly lower than in 2016 (χ²=12.05, P<0.01), 2017 (χ²=7.34, P=0.01) and 2018 (χ²=6.78, P=0.01), but there was no significant difference compared with 2015 (χ²=2.84, P=0.09). The HCoV-positive rate of 2020 (3.37%) was significantly higher than in 2015 (χ²=13.636, P<0.01), 2016 (χ²=11.099, P<0.01), 2017 (χ²=15.482, P<0.01), 2018(χ²=18.601, P<0.01) and 2019(χ²=45.580, P<0.01). The positive rate was highest in spring (March to May) in 2015 and 2017 to 2018. February to April and July to September of 2016 were the peak periods of positive detection. No obvious seasonal change was observed in 2019 and the HCoV-positive rate of 2020 was extremely low from January to July, following significantly increased from August to December. 26.37% (280/1 062) of HCoV were co-detected with other respiratory pathogens and the most frequently identified mixed detection was RSV. Three or more pathogens were detected in 7.34% (78/1 062) of the HCoV-positive samples. Bronchopneumonia and bronchiolitis were more frequently observed in the single HCoV positive (61.89% and 16.75%) children compared to co-detected children(34.29% and 9.64%)(χ²=63.394 and 8.228, P<0.01). However, compared to those with HCoV mono-detection, co-detected children were more likely to have severe pneumonia (4.6% and 47.14%) (χ²=280.171, P<0.01). Conclusions: HCoV is one of the respiratory pathogens in children in Hebei region and more prevalent in spring. The susceptible population of HCoV is mainly children under the age of 3 years old. HCoV often co-detects with other respiratory pathogens, and the co-infection is one of the risk factors of severe pneumonia in children with respiratory infection.

Hybrid analysis for understanding contact tracing apps adoption

Purpose This study aims to explore the adoption of contact tracing apps through a hybrid analysis of the collected data using structural equation modelling (SEM) and artificial neural networks (ANN), leading to the identification of the critical determinants for the adoption of contact tracing apps in Australia. Design/methodology/approach A research model is developed within the background of the unified theory of acceptance and use of technology (UTAUT) and the privacy calculus theory (PCT) for investigating the adoption of contact tracing apps. This model is then tested and validated using a hybrid SEM-ANN analysis of the survey data. Findings The study shows that effort expectancy, perceived value of information disclosure and social influence are critical for adopting contact tracing apps. It reveals that performance expectancy and perceived privacy risks are indirectly significant on the adoption through the influence of perceived value of information disclosure. Furthermore, the study finds out that facilitating condition is insignificant to the adoption of contact tracing apps. Practical implications The findings of the study can lead to the formulation of targeted strategies and policies for promoting the adoption of contact tracing apps and inform future epidemic control for better emergency management. Originality/value This study is the first attempt in integrating UTAUT and PCT for exploring the adoption of contact tracing apps in Australia. It combines SEM and ANN for analysing the survey data, leading to better understanding of the critical determinants for the adoption of contact tracing apps.

The influence of the corporate social responsibility disclosures on consumer brand attitudes under the impact of COVID-19

This study focuses on the use of we-media by small- and medium-sized enterprises (SMEs) to disclose internal corporate social responsibility (ICSR) under the impact of the 2019 novel coronavirus disease (COVID-19). Study 1 interprets the catalyst effect of COVID-19 on the externalization of SMEs’ ICSR. The fuzzy grading evaluation method is initially verified. Under the impact of COVID-19, SMEs fulfilling their ICSR can enhance consumer brand attitudes. Study 2 uses a structural equation model and empirical analysis of 946 effective samples and finds that consumers perceive the self-sacrifice of corporations during the coronavirus disease period. SMEs can fulfill their ICSR to enhance the internal explanation mechanism of consumer brand attitudes and the moderating role of enterprise losses. © 2020, The Author(s).