MedConQA: Medical Conversational Question Answering System based on Knowledge Graphs

The medical conversational system can relieve doctors' burden and improve healthcare effi-ciency, especially during the COVID-19 pan-demic. However, the existing medical dialogue systems have die problems of weak scalability, insufficient knowledge, and poor controlla-bility. Thus, we propose a medical conversa-tional question-answering (CQA) system based on the knowledge graph, namely MedConQA, which is designed as a pipeline framework to maintain high flexibility. Our system utilizes automated medical procedures, including medi-cal triage, consultation, image-text drug recom-mendation, and record. Each module has been open-sourced as a tool, which can be used alone or in combination, with robust scalability. Besides, to conduct knowledge-grounded dia-logues with users, we first construct a Chinese Medical Knowledge Graph (CMKG) and col-lect a large-scale Chinese Medical CQA (CM-CQA) dataset, and we design a series of meth-ods for reasoning more intellectually. Finally, we use several state-of-the-art (SOTA) tech-niques to keep the final generated response more controllable, which is further assured by hospital and professional evaluations. We have open-sourced related code, datasets, web pages, and tools, hoping to advance future research. © 2022 Association for Computational Linguistics.

PANDORA: Deep Graph Learning Based COVID-19 Infection Risk Level Forecasting

Coronavirus disease 2019 (COVID-19) as a global pandemic causes a massive disruption to social stability that threatens human life and the economy. An effective forecasting system is arguably important to provide an early signal of the risk of COVID-19 infection so that the authorities are ready to protect the people from the worst. However, making a good forecasting model for infection risks in different cities or regions is not an easy task, because it has a lot of influential factors that are difficult to be identified manually. To address the current limitations, we propose a deep graph learning model, called PANDORA, to predict the infection risks of COVID-19, by considering all essential factors and integrating them into a geographical network. The framework uses geographical position relationships and transportation frequency as higher order structural properties formulated by higher order network structures (i.e., network motifs). Moreover, four significant node attributes (i.e., multiple features of a particular area, including climate, medical condition, economy, and human mobility) are also considered. We propose three different aggregators to better aggregate node attributes and structural features, namely, Hadamard, Summation, and Connection. Experimental results over real data show that PANDORA outperforms the baseline methods with higher accuracy and faster convergence speed, no matter which aggregator is chosen. IEEE

COVID-19 Datasets: A Brief Overview

The outbreak of the COVID-19 pandemic affects lives and social-economic development around the world. The affecting of the pandemic has motivated researchers from different domains to find effective solutions to diagnose, prevent, and estimate the pandemic and relieve its adverse effects. Numerous COVID-19 datasets are built from these studies and are available to the public. These datasets can be used for disease diagnosis and case prediction, speeding up solving problems caused by the pandemic. To meet the needs of researchers to understand various COVID-19 datasets, we examine and provide an overview of them. We organise the majority of these datasets into three categories based on the category of ap-plications, i.e., time-series, knowledge base, and media-based datasets. Organising COVID-19 datasets into appropriate categories can help researchers hold their focus on methodology rather than the datasets. In addition, applications and COVID-19 datasets suffer from a series of problems, such as privacy and quality. We discuss these issues as well as potentials of COVID-19 datasets. © 2022, ComSIS Consortium. All rights reserved.

Exploring Public Sentiment During COVID-19: A Cross Country Analysis

COVID-19 has spread all over the world, accounting for countless death and enormous economic loss. Since the World Health Organization (WHO) declared COVID-19 as a pandemic, governments from different countries have made various policies to prevent the pandemic from becoming worse. However, civilian reactions to the pandemic vary when they face similar situations. This behavioral variation creates a challenge when it comes to policy-making. Such differences are generally implicit, hidden in ones' social lives. As a result, it is challenging to analyze such differences when the governments make policies. In this work, we investigate social media posts on Twitter and Weibo in order to effectively explore the difference in reactions across various countries, with the aim to understand national differences. To this end, we employ natural language processing (NLP) methods and Linguistic Inquiry and Word Count (LIWC) tools to process six languages in different countries, including the USA, Germany, France, Italy, the U.K., and China. We provide a comprehensive analysis of public reaction differences from the emotional perspective. Our findings verify that the reactions vary noticeably among various countries for some policies. Therefore, sentiment analysis can significantly influence policy-making. Our work sheds light on the mechanism of detecting the reaction differences in various countries, which can be utilized to conduct effective communication and make appropriate policy decisions.

Effect of Online Interaction on College Student Satisfaction in Online Courses: A Chained Mediation Model

The large-scale online teaching during the COVID-19 epidemic is different from the past, as remote real-time synchronous interaction has been realized in most regions. Under the new situation, this study investigated how interactions affect student satisfaction in online courses considering the correlation between interactive external factors and individual psychological factors based on the constructivist learning theory. A total of 2,685 undergraduate students who had online learning experience during the COVID-19 epidemic were surveyed. Structural equation modeling was used to examine the direct link between instructional interaction and student satisfaction in online courses as well as indirect links through task value and online learning self-efficacy. Results indicate as this: (1) Online interaction contributes to student satisfaction both directly and indirectly, and the total indirect effect is greater than the direct one. (2) Among the indirect effects, online interaction can enhance student satisfaction both through the sequential chain mediating effect of task value and self-efficacy and through the separate mediating effect of task value and self-efficacy. Results suggest that great importance should be attached to the role of psychological factors such as task value and self-efficacy when constructing interactive mode in online courses. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Data-Driven Decision-Making in COVID-19 Response: A Survey

COVID-19 has spread all over the world, having an enormous effect on our daily life and work. In response to the epidemic, a lot of important decisions need to be taken to save communities and economies worldwide. Data clearly play a vital role in effective decision-making. Data-driven decision-making uses data-related evidence and insights to guide the decision-making process and verify the plan of action before it is committed. To better handle the epidemic, governments and policy-making institutes have investigated abundant data originating from COVID-19. These data include those related to medicine, knowledge, media, and so on. Based on these data, many prevention and control policies are made. In this survey article, we summarize the progress of data-driven decision-making in the response to COVID-19, including COVID-19 prevention and control, psychological counseling, financial aid, work resumption, and school reopening. We also propose some current challenges and open issues in data-driven decision-making, including data collection and quality, complex data analysis, and fairness in decision-making. This survey article sheds light on current policy-making driven by data, which also provides a feasible direction for further scientific research. IEEE

Evaluating factors mediating the relationship between male sex and COVID-19 severity

Background: Males have experienced higher rates of severe COVID-19 outcomes compared to females but the underlying causal mechanisms of this relationship are not well understood. We leveraged existing electronic medical records (EMR) to evaluate associations between sex and COVID-19 test positivity, disease severity, viral burden, and death, and assess factors that mediate the relationship between male sex and severe COVID-19 disease. Methods: We conducted a retrospective cohort study with data collected from University of Washington Medicine EMR from March 1 to September 29, 2020. All persons, regardless of age, were included if they had a conclusive diagnostic COVID-19 PCR test result. We defined severe COVID-19 disease as a score >5 on the WHO clinical progression scale. We used Poisson regression to assess sex differences in risk for COVID-19 test positivity, disease severity and COVID-19 related death, and linear regression to compare viral cycle threshold at the first positive test. We conducted mediation analyses to assess interventional indirect effects of male sex on severe COVID-19 risk through socioeconomic status (SES, based on area deprivation and insurance type), comorbidities, and inflammation status. Models controlled for age and race/ethnicity. Results: Of individuals with SARS-CoV-2 testing records, 32,919 males and 34,733 females had a conclusive PCR test during our observation period. Males were 13% more likely to test positive than females in multivariable analysis (RR=1.13;95% CI: 1.04-1.24;Table). Males had 85% higher risk for severe COVID-19 disease (RR=1.85;95% CI: 1.33-2.62) and 66% higher risk for COVID-19 related death (RR=1.66;95% CI: 0.95-2.98) than females following a positive test result. No difference was observed in cycle threshold at first positive test between males and females (p=0.69). Mediation analyses indicated a significant interventional indirect effect of male sex on severe COVID-19 disease through inflammation status (RR=1.07;95% CI: 1.01-1.13), and less so through SES or comorbidities. Conclusion: In our cohort, males had higher test positivity and greater risk of COVID-19 severity and death. This relationship between male sex and severe COVID-19 seems to act in part through inflammation status. Additional analyses in larger cohorts are needed to better understand the full range of socio-behavioral and biologic factors that mediate the relationship between sex and poor COVID-19 outcomes. (Figure Presented).

CRI: Measuring City Infection Risk amid COVID-19

The outbreak of COVID-19 has brought incalculable economy and life losses. Accurately assessing the risk of a certain city can help formulate effective measures to prevent and control COVID-19 in time. It will be of great significance for us to measure city risk in infection amid epidemics. City risk in infection is related to many factors. To address this problem, this paper proposes city risk index (CRI) to measure city risk in infection, considering the following four perspectives: Economy (i.e., GDP and FCI), technology (i.e., education and innovation), population, and geographical position (i.e., latitude and longitude). The experimental results show that CRI can be effectively employed to measure city risk in infection amid COVID-19 as well as other similar epidemics. The proposed CRI can be used to guide policymakers for better emergency management policies making when coping with COVID-19. © 2020 IEEE.

441 Cutaneous findings in COVID-19 patients hospitalized at a large urban academic medical center

Background: Cutaneous manifestations have been associated with COVID-19 infection and their clinical significance in hospitalized patients remains unclear. Methods: A retrospective chart review of 1216 patients older than 18 years of age hospitalized with laboratory-confirmed SARS-CoV-2 infection from March 12, 2020 to May 31, 2020 at a large urban academic medical center. A keyword search query of patient records combined with manual chart review by at least two dermatologists identified a study group having cutaneous manifestations concurrent with COVID-19 infection, specifically between 14 days prior to admission and up to discharge. Results: 122 patients with 195 skin lesions concurrent with COVID-19 hospitalization were identified. Dermatology reviewers evaluated clinical photographs for 116 lesions (59.5%) and inpatient dermatology consultations for 42 lesions (21.5%). The most common cutaneous findings in patients with COVID-19 hospitalization were pressure injuries (n=118;60.5%) and morbilliform eruptions (n=33;16.9%). A very small number of patients (0.6%;n=7/1216) had exanthems occurring within 2 weeks of COVID-19 symptom onset. The majority of exanthems developed within 14 days of exposure to possible culprit drugs and beyond the 14-day window of COVID-19 symptom onset, making viral association unlikely. Conclusion: Skin lesions concurrent with COVID-19 hospitalization were most frequently linked to hospitalization-related factors, such as pressure injuries or drug-related exanthems, rather than due to novel pathologies related to SARS-CoV-2 itself.

Clinical and radiological characteristics of COVID-19: a multicentre, retrospective, observational study.

BACKGROUND: Multicentre cohort investigations of patients with coronavirus disease 2019 (COVID-19) have been limited. We investigated the clinical and chest computed tomography characteristics of patients with COVID-19 at the peak of the epidemic from multiple centres in China. METHODS: We retrospectively analysed the epidemiologic, clinical, laboratory, and radiological characteristics of 189 patients with confirmed COVID-19 who were admitted to seven hospitals in four Chinese provinces from 18 January 2020 to 3 February 2020. RESULTS: The mean patient age was 44 years and 52.9% were men; 186/189 had ≥1 co-existing medical condition. Fever, cough, fatigue, myalgia, diarrhoea, and headache were common symptoms at onset; hypertension was the most common co-morbidity. Common clinical signs included dyspnoea, hypoxia, leukopenia, lymphocytopenia, and neutropenia; most lesions exhibited subpleural distribution. The most common radiological manifestation was mixed ground-glass opacity with consolidation (mGGO-C); most patients had grid-like shadows and some showed paving stones. Patients with hypertension, dyspnoea, or hypoxia exhibited more severe lobe involvement and diffusely distributed lesions. Patients in severely affected areas exhibited higher body temperature; more fatigue and dyspnoea; and more manifestations of multiple lesions, lobe involvement, and mGGO-C. During the Wuhan lockdown period, cough, nausea, and dyspnoea were alleviated in patients with newly confirmed COVID-19; lobe involvement was also improved. CONCLUSIONS: Among patients with COVID-19 hospitalised at the peak of the epidemic in China, fever, cough, and dyspnoea were the main symptoms at initial diagnosis, accompanied by lymphocytopenia and hypoxaemia. Patients with severe disease showed more severe lobe involvement and diffuse pulmonary lesion distribution.