Unsupervised Domain Adaptation for COVID-19 Information Service with Contrastive Adversarial Domain Mixup

In the real-world application of COVID-19 misinformation detection, a fundamental challenge is the lack of the labeled COVID data to enable supervised end-to-end training of the models, especially at the early stage of the pandemic. To address this challenge, we propose an unsupervised domain adaptation framework using contrastive learning and adversarial domain mixup to transfer the knowledge from an existing source data domain to the target COVID-19 data domain. In particular, to bridge the gap between the source domain and the target domain, our method reduces a radial basis function (RBF) based discrepancy between these two domains. Moreover, we leverage the power of domain adversarial examples to establish an intermediate domain mixup, where the latent representations of the input text from both domains could be mixed during the training process. Extensive experiments on multiple real-world datasets suggest that our method can effectively adapt misinformation detection systems to the unseen COVID-19 target domain with significant improvements compared to the state-of-the-art baselines. © 2022 IEEE.

A Knowledge-driven Domain Adaptive Approach to Early Misinformation Detection in an Emergent Health Domain on Social Media

This paper focuses on an important problem of early misinformation detection in an emergent health domain on social media. Current misinformation detection solutions often suffer from the lack of resources (e.g., labeled datasets, sufficient medical knowledge) in the emerging health domain to accurately identify online misinformation at an early stage. To address such a limitation, we develop a knowledge-driven domain adaptive approach that explores a good set of annotated data and reliable knowledge facts in a source domain (e.g., COVID-19) to learn the domain-invariant features that can be adapted to detect misinformation in the emergent target domain with little ground truth labels (e.g., Monkeypox). Two critical challenges exist in developing our solution: i) how to leverage the noisy knowledge facts in the source domain to obtain the medical knowledge related to the target domain? ii) How to adapt the domain discrepancy between the source and target domains to accurately assess the truthfulness of the social media posts in the target domain? To address the above challenges, we develop KAdapt, a knowledge-driven domain adaptive early misinformation detection framework that explicitly extracts rel-evant knowledge facts from the source domain and jointly learns the domain-invariant representation of the social media posts and their relevant knowledge facts to accurately identify misleading posts in the target domain. Evaluation results on five real-world datasets demonstrate that KAdapt significantly outperforms state-of-the-art baselines in terms of accurately detecting misleading Monkeypox posts on social media. © 2022 IEEE.

Contrastive Domain Adaptation for Early Misinformation Detection: A Case Study on COVID-19

Despite recent progress in improving the performance of misinformation detection systems, classifying misinformation in an unseen domain remains an elusive challenge. To address this issue, a common approach is to introduce a domain critic and encourage domain-invariant input features. However, early misinformation often demonstrates both conditional and label shifts against existing misinformation data (e.g., class imbalance in COVID-19 datasets), rendering such methods less effective for detecting early misinformation. In this paper, we propose contrastive adaptation network for early misinformation detection (CANMD). Specifically, we leverage pseudo labeling to generate high-confidence target examples for joint training with source data. We additionally design a label correction component to estimate and correct the label shifts (i.e., class priors) between the source and target domains. Moreover, a contrastive adaptation loss is integrated in the objective function to reduce the intra-class discrepancy and enlarge the inter-class discrepancy. As such, the adapted model learns corrected class priors and an invariant conditional distribution across both domains for improved estimation of the target data distribution. To demonstrate the effectiveness of the proposed CANMD, we study the case of COVID-19 early misinformation detection and perform extensive experiments using multiple real-world datasets. The results suggest that CANMD can effectively adapt misinformation detection systems to the unseen COVID-19 target domain with significant improvements compared to the state-of-the-art baselines. © 2022 ACM.

Flexible Solar-Blind Photodetectors Based on β-Ga2O3 Films Transferred by a Stamp-Based Printing Technique

A stamp-based printing technique was applied to transfer the β-Ga₂O₃ films grown by pulsed laser deposition (PLD) from Si substrates onto some flexible substrates, such as PET, PEN, and PI. It is demonstrated that the β-Ga₂O₃-based flexible solar-blind photodetectors (SBPDs) exhibit brilliant optoelectrical performances with a low dark current of 1.7 pA at 10 V, a I_254nm/I_dark ratio of 1.2×103, rise (τ_r1 = 0.079 s and τ_r2 = 0.413 s) and decay (τ_d1 = 0.029 s and τ_d2 = 0.316 s) times. In a further step, flexible imaging sensor arrays based on the β-Ga₂O₃/PET were fabricated, which exhibit good imaging capability and resolution. Moreover, wearable UVC-alarms based on the β-Ga₂O₃/PET were realized to monitor the UVC radiation in the environment in real time, which can be used in the COVID-19-related area. IEEE

A Privacy-aware Distributed Knowledge Graph Approach to QoIS-driven COVID-19 Misinformation Detection

In this paper, we focus on the quality of information service (QoIS) of COVID-19-related information on social media. Our goal is to provide reliable COVID-19 information service by accurately detecting the misleading COVID-19 posts on social media by exploring the community-contributed COVID-19 fact data (CCFD) from different social media platforms. In particular, CCFD refers to the fact-checking reports that are submitted to each social media platform by its users and fact-checking professionals. Our work is motivated by the observation that CCFD often contains useful COVID-19 knowledge facts (e.g., "COVID-19 is not a flu") that can effectively facilitate the identification of misleading COVID-19 social media posts. However, CCFD is often private to the individual social media platform that owns it due to the data privacy concerns such as data copyright of CCFD and user profile information of CCFD contributors. In this paper, we leverage the CCFD from different social media platforms to accurately detect COVID19 misinformation while effectively protecting the privacy of CCFD. Two critical challenges exist in solving our problem: 1) how to generate privacy-aware COVID-19 knowledge facts from the platform-specific CCFD? 2) How to effectively integrate the privacy-aware COVID-19 knowledge facts from different social media platforms to correctly assess the truthfulness of a COVID19 post? To address these challenges, we develop CoviDKG, a COVID-19 distributed knowledge graph framework that constructs a set of CCFD-based knowledge graphs on individual social media platform and exchanges the privacy-aware COVID19 knowledge facts across different platforms to effectively detect misleading COVID-19 posts. We evaluate CoviDKG on two real-world social media datasets and the results show that CoviDKG achieves significant performance gains compared to state-of-the-art baselines in accurately detecting misleading COVID-19 posts on social media. © 2022 IEEE.

A Duo-generative Approach to Explainable Multimodal COVID-19 Misinformation Detection

This paper focuses on a critical problem of explainable multimodal COVID-19 misinformation detection where the goal is to accurately detect misleading information in multimodal COVID-19 news articles and provide the reason or evidence that can explain the detection results. Our work is motivated by the lack of judicious study of the association between different modalities (e.g., text and image) of the COVID-19 news content in current solutions. In this paper, we present a generative approach to detect multimodal COVID-19 misinformation by investigating the cross-modal association between the visual and textual content that is deeply embedded in the multimodal news content. Two critical challenges exist in developing our solution: 1) how to accurately assess the consistency between the visual and textual content of a multimodal COVID-19 news article? 2) How to effectively retrieve useful information from the unreliable user comments to explain the misinformation detection results? To address the above challenges, we develop a duo-generative explainable misinformation detection (DGExplain) framework that explicitly explores the cross-modal association between the news content in different modalities and effectively exploits user comments to detect and explain misinformation in multimodal COVID-19 news articles. We evaluate DGExplain on two real-world multimodal COVID-19 news datasets. Evaluation results demonstrate that DGExplain significantly outperforms state-of-the-art baselines in terms of the accuracy of multimodal COVID-19 misinformation detection and the explainability of detection explanations. © 2022 ACM.

A Multimodal Misinformation Detector for COVID-19 Short Videos on TikTok

This paper studies an emerging and important problem of identifying misleading COVID-19 short videos where the misleading content is jointly expressed in the visual, audio, and textual content of videos. Existing solutions for misleading video detection mainly focus on the authenticity of videos or audios against AI algorithms (e.g., deepfake) or video manipulation, and are insufficient to address our problem where most videos are user-generated and intentionally edited. Two critical challenges exist in solving our problem: i) how to effectively extract information from the distractive and manipulated visual content in TikTok videos? ii) How to efficiently aggregate heterogeneous information across different modalities in short videos? To address the above challenges, we develop TikTec, a multimodal misinformation detection framework that explicitly exploits the captions to accurately capture the key information from the distractive video content, and effectively learns the composed misinformation that is jointly conveyed by the visual and audio content. We evaluate TikTec on a real-world COVID- 19 video dataset collected from TikTok. Evaluation results show that TikTec achieves significant performance gains compared to state-of-the-art baselines in accurately detecting misleading COVID-19 short videos. © 2021 IEEE.

Traditional Chinese medicine together with high-dose vitamin C improves the therapeutic effect of western medicine against COVID-19

Objectives: Traditional Chinese medicine has been reported to be effective in the treatment of epidemic diseases. Here, we aimed to investigate the effects of combined therapy of Chinese and western medicine on coronavirus disease 2019 (COVID-19). Methods: A total of 60 patients diagnosed with COVID-19 were enrolled. Both the ordinary and severely affected patients were randomly divided into Groups A-C each with 10 cases each. The patients in Group A-C received Western medicine, Western medicine + traditional Chinese medicine, and Western medicine + traditional Chinese medicine + high dose of vitamin C, respectively. The time of disease recovery, symptoms disappearance, chest CT improvement, and tongue amelioration was recorded. Leukocyte, neutrophil and lymphocyte were monitored, as well as C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), procalitonin (PCT), inflammatory factors, partial pressure of oxygen and carbon dioxide (PaCO2) and oxygenation index (PaO2). Urinary tract stones, liver function, and other side-effects such as gastrointestinal dysfunction were also investigated. Results: Traditional Chinese medicine enhanced the effect of Western medicine, including the reduction of CRP, ESR, PCT, and inflammatory factors, and the increase of leukocyte, neutrophil, and lymphocyte counts, and the improvement of respiratory rate, PaO2, PaCO2, and oxygenation index. Traditional Chinese medicine combined with high-dose Vitamin C therapy more effectively shortened the time of disease recovery, symptom disappearance, chest CT improvement, and tongue amelioration. Conclusions: a combined therapy of Western medicine, traditional Chinese medicine, and high dose of Vitamin C results in a most effective outcome in the treatment of COVID-19.

Clinical effect of the treatment of novel coronavirus pneumonia by internal administration of traditional Chinese medicine plus fumigation and absorption combined with super dose of vitamin C in treating COVID-19

Objective: To study the characteristics and regularity of the improvement of early clinical symptoms of coronavirus disease 2019 (COVID-19) treated with Chinese medicine plus fumigation and absorption combined with super dose of vitamin C. Methods: We randomly divided 30 patients diagnosed with COVID-19 admitted by the Hubei medical team in our hospital since February 2020 into groups A, B and C, with 10 cases in each group. Group A was the control group. Group B was treated with traditional Chinese medicine and fumigation. Group C was the treatment group of traditional Chinese medicine plus fumigation and absorption combined with super dose of vitamin C. We analyzed the symptoms of fatigue, cough, dry throat, shortness of breath and the improvement of chest CT and nucleic acid detection, and compared the treatment status of each group. Results: The improvement degree and disappearance time of fatigue, cough, dry throat and shortness of breath in group B and group C were better than those in group A, and the effect of group C was better than that of group B (P<0.05 or P<0.01). No significant statistical difference was found in chest CT scanning or nucleic acid detection results (P>0.05). Conclusion: The combination of traditional Chinese medicine and fumigation and absorption combined with super dose of vitamin C has a definite effect on the improvement of fatigue, cough, dry throat and shortness of breath in patients with COVID-19. © 2020, Editorial Board of Journal of Xi'an Jiaotong University (Medical Sciences). All right reserved.

The influencing factors of interprovincial transmission and development of COVID-19: Data analysis based on 30 provinces and cities in China

Objective: To explore the factors affecting the interprovincial transmission and development of coronavirus disease 2019 (COVID-19) in China, with a view to providing recommendations for the formulation of preventive and control measures according to the actual conditions in different regions during the outbreak of the severe infectious disease. Methods: We collected the total number of confirmed cases of COVID-19 in 30 provinces and cities in China by the end of 24:00 February 25, 2020. Then we also collected the distance from each region to Hubei province, the proportion of population moving out from Wuhan city from January 1 to January 23, population density, urban population, traffic passenger volume, passenger turnover volume and other relevant data of each region.The cumulative confirmed cases including the most of imported cases by the end of 24:00 January 29, 2020 were taken as the first-stage cases cluster, and the cumulative newly confirmed cases including the most of secondary cases from 0:00 January 30 to 24:00 February 25, 2020 were taken as the second-stage cases cluster. Pearson bivariate correlation and linear fitting regression method were adopted to analyze the effects of population migration, transportation, economy and other factors on the transmission and development of COVID-19 in different regions. In the linear fitting regression, the multi-factor optimal subset model was used to screen the factors most closely related to COVID-19. Results: The distance from each region to Hubei province was negatively correlated with the first-stage cases cluster with the most of imported cases and the second-stage cases cluster with the most of secondary cases(t=-3.654, t=-3.679, both P<0.05). The proportion of population moving out from Wuhan, GDP, urban population, traffic passenger volume, and passenger turnover volume were positively correlated with the first-stage and second-stage cases cluster (all t>2.760, all P<0.05). GDP and the proportion of population moving out from Wuhan were most closely related to the first-stage cases cluster with the most of imported cases (t=4.173, t=7.851, all P<0.05). The first-stage cases cluster, the proportion of population moving out from Wuhan, and urban population were most closely related to the second-stage cases cluster with the most of secondary cases (t=4.734, t=3.491, t=2.855, all P<0.05). Results: GDP and the proportion of population moving out from Wuhan city had the greatest impact on the stage with the most of imported cases. The imported cases, the proportion of population moving out from Wuhan and the urban population had the greatest impact on the stage with the most of secondary cases. In the early stage of epidemic outbreak with the most of imported cases,we should consider strengthening the prevention and control of the epidemic in areas with high level of GDP and high proportion of population moving out from the epidemic area.The flow of population should be restricted more strictly as soon as possible in order to effectively curb the outbreak of the epidemic.In the later-stage of epidemic with the most of secondary cases, regionalized control policies should be formulated mainly according to the indicators of imported cases, the population proportion fromtheepidemic area, and the urban population. Finally, the contact of population should be restricted reasonably to prevent further development of the epidemic.