Analyzing Impact of Socio-Economic Factors on COVID-19 Mortality Prediction Using SHAP Value

This paper applies multiple machine learning (ML) algorithms to a dataset of de-identified COVID-19 patients provided by the COVID-19 Research Database. The dataset consists of 20,878 COVID-positive patients, among which 9,177 patients died in the year 2020. This paper aims to understand and interpret the association of socio-economic characteristics of patients with their mortality instead of maximizing prediction accuracy. According to our analysis, a patient's household's annual and disposable income, age, education, and employment status significantly impacts a machine learning model's prediction. We also observe several individual patient data, which gives us insight into how the feature values impact the prediction for that data point. This paper analyzes the global and local interpretation of machine learning models on socio-economic data of COVID patients.

Psychological recovery effects of 3D virtual tourism with real scenes -- a comparative study

Objective This study explores the psychological recovery effects of virtual tourism on individuals. Methods Relevant research usually tends to examine the psychological recovery effects through traditional media entailing a lesser immersive experience. Few studies focus on the psychological recovery effects of virtual tourism, and even fewer on exploring response differences depending on different landscape types. Based on a series of empirical tests and electroencephalogram (EEG) data, this study investigates the impacts of a more immersive 3D virtual tourism with real scenes on people's relaxation (Pm), concentration (Pa), and positive and negative emotions (PA and NA). Additionally, it clarifies the differences in the psychological recovery effects of four landscape types on the abovementioned attributes. Conclusions The psychological recovery effects did vary according to the type of tourist attractions. There were a few differences based on gender. For instance, men's relaxation level changed significantly after touring lake-oriented virtual tourist attraction. Individual differences in recovery were also observed. Implications These findings contribute to our knowledge about environmental restoration and its role in alleviating people's anxiety, especially during situations like the COVID-19 pandemic.

Detrimental impact of contagious disease cues on consumer preference for anthropomorphic products.

Despite the extensive use of anthropomorphism strategy in marketing practices, little research attention has been given to the environmental factors that influence consumer preference for anthropomorphic products. This research examines when and why contagious disease cues can influence consumer preference for anthropomorphic products. The results from four empirical experiments consistently show that when exposed to contagious disease cues, consumers exhibit a lower preference for anthropomorphic products (Study 1), which is mediated by social withdrawal (Study 2). Furthermore, our findings demonstrate that this detrimental effect would be attenuated for products in digital (vs. physical) format (Study 3), or in regions with low (vs. high) local severity of the contagious disease (Study 4). These findings contribute to the literature on contagious diseases and anthropomorphism and offer important managerial implications. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1007/s11002-022-09614-x.

Lower vaccine-acquired immunity in the elderly population following two-dose BNT162b2 vaccination is alleviated by a third vaccine dose.

Understanding the impact of age on vaccinations is essential for the design and delivery of vaccines against SARS-CoV-2. Here, we present findings from a comprehensive analysis of multiple compartments of the memory immune response in 312 individuals vaccinated with the BNT162b2 SARS-CoV-2 mRNA vaccine. Two vaccine doses induce high antibody and T cell responses in most individuals. However, antibody recognition of the Spike protein of the Delta and Omicron variants is less efficient than that of the ancestral Wuhan strain. Age-stratified analyses identify a group of low antibody responders where individuals ≥60 years are overrepresented. Waning of the antibody and cellular responses is observed in 30% of the vaccinees after 6 months. However, age does not influence the waning of these responses. Taken together, while individuals ≥60 years old take longer to acquire vaccine-induced immunity, they develop more sustained acquired immunity at 6 months post-vaccination. A third dose strongly boosts the low antibody responses in the older individuals against the ancestral Wuhan strain, Delta and Omicron variants.

Supervised Pretraining through Contrastive Categorical Positive Samplings to Improve COVID-19 Mortality Prediction.

Clinical EHR data is naturally heterogeneous, where it contains abundant sub-phenotype. Such diversity creates challenges for outcome prediction using a machine learning model since it leads to high intra-class variance. To address this issue, we propose a supervised pre-training model with a unique embedded k-nearest-neighbor positive sampling strategy. We demonstrate the enhanced performance value of this framework theoretically and show that it yields highly competitive experimental results in predicting patient mortality in real-world COVID-19 EHR data with a total of over 7,000 patients admitted to a large, urban health system. Our method achieves a better AUROC prediction score of 0.872, which outperforms the alternative pre-training models and traditional machine learning methods. Additionally, our method performs much better when the training data size is small (345 training instances).

[Clinical features of children with coronavirus disease 2019 Delta variant infection after vaccination with inactivated SARS-CoV-2 vaccine]. / æŽ¥ç§æ–°åž‹å† çŠ¶ç— æ¯’ç­æ´»ç–«è‹—åŽæ„ŸæŸ“Deltaå˜å¼‚æ ªçš„æ–°åž‹å† çŠ¶ç— æ¯’è‚ºç‚Žæ‚£å„¿ä¸´åºŠç‰¹å¾åˆ†æž.

OBJECTIVES: To study the clinical features of children with coronavirus disease 2019 (COVID-19) Delta variant infection vaccinated or not vaccinated with inactivated severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccine. METHODS: A total of 11 children with COVID-19 Delta variant infection who were vaccinated with inactivated SARS-CoV-2 vaccine and were hospitalized in the designated hospital in Henan Province, China, from November 3 to December 17, 2021 were enrolled as the vaccinated group. Thirty-one children with COVID-19 Delta variant infection who were not vaccinated and were hospitalized during the same period were enrolled as the unvaccinated group. A retrospective analysis was performed on their epidemiological data, clinical features, and laboratory examination results. RESULTS: There was no significant difference in gender composition and disease classification between the two groups (P>0.05), and there was also no significant difference in the incidence rates of the clinical symptoms such as cough, expectoration, and fever between the two groups (P>0.05). No significant difference was found between the two groups in leukocyte count, lymphocyte percentage, alanine aminotransferase, and serum creatinine (P>0.05). Compared with the unvaccinated group, the vaccinated group had significantly lower levels of aspartate aminotransferase, lactate dehydrogenase, and creatine kinase-MB (P<0.05). There was no significant difference between the two groups in the proportion of children with elevated C-reactive protein or procalcitonin and the levels of peripheral blood cytokines (P>0.05). The vaccinated group had significantly lower counts of B lymphocytes and total T lymphocytes (CD3+) than the unvaccinated group (P<0.05). Compared with the unvaccinated group, the vaccinated group had a significantly higher positive rate of IgG on admission and at week 2 of the course of disease (P<0.05), as well as a significantly higher Ct value of nucleic acid at weeks 1 and 2 of the course of disease (P<0.05). CONCLUSIONS: Vaccination with inactivated SARS-CoV-2 vaccine may reduce myocardial injury caused by SARS-CoV-2 Delta variant. For children with SARS-CoV-2 Delta variant infection after the vaccination, more attention should be paid to their immune function.

Further divided gender gaps in research productivity and collaboration during the COVID-19 pandemic: Evidence from coronavirus-related literature.

Based on publication data on coronavirus-related fields, this study applies a difference in differences approach to explore the evolution of gender inequalities before and during the COVID-19 pandemic by comparing the differences in the numbers and shares of authorships, leadership in publications, gender composition of collaboration, and scientific impacts. We find that, during the pandemic: (1) females' leadership in publications as the first author was negatively affected; (2) although both females and males published more papers relative to the pre-pandemic period, the gender gaps in the share of authorships have been strengthened due to the larger increase in males' authorships; (3) the share of publications by mixed-gender collaboration declined; (4) papers by teams in which females play a key role were less cited in the pre-pandemic period, and this citation disadvantage was exacerbated during the pandemic; and (5) gender inequalities regarding authorships and collaboration were enhanced in the initial stage of COVID-19, widened with the increasing severity of COVID-19, and returned to the pre-pandemic level in September 2020. This study shows that females' lower participation in teams as major contributors and less collaboration with their male colleagues also reflect their underrepresentation in science in the pandemic period. This investigation significantly deepens our understanding of how the pandemic influenced academia, based on which science policies and gender policy changes are proposed to mitigate the gender gaps.

Detrimental impact of contagious disease cues on consumer preference for anthropomorphic products

Despite the extensive use of anthropomorphism strategy in marketing practices, little research attention has been given to the environmental factors that influence consumer preference for anthropomorphic products. This research examines when and why contagious disease cues can influence consumer preference for anthropomorphic products. The results from four empirical experiments consistently show that when exposed to contagious disease cues, consumers exhibit a lower preference for anthropomorphic products (Study 1), which is mediated by social withdrawal (Study 2). Furthermore, our findings demonstrate that this detrimental effect would be attenuated for products in digital (vs. physical) format (Study 3), or in regions with low (vs. high) local severity of the contagious disease (Study 4). These findings contribute to the literature on contagious diseases and anthropomorphism and offer important managerial implications. Supplementary Information The online version contains supplementary material available at 10.1007/s11002-022-09614-x.

Pandemics are catalysts of scientific novelty: Evidence from COVID-19.

Scientific novelty drives the efforts to invent new vaccines and solutions during the pandemic. First-time collaboration and international collaboration are two pivotal channels to expand teams' search activities for a broader scope of resources required to address the global challenge, which might facilitate the generation of novel ideas. Our analysis of 98,981 coronavirus papers suggests that scientific novelty measured by the BioBERT model that is pretrained on 29 million PubMed articles, and first-time collaboration increased after the outbreak of COVID-19, and international collaboration witnessed a sudden decrease. During COVID-19, papers with more first-time collaboration were found to be more novel and international collaboration did not hamper novelty as it had done in the normal periods. The findings suggest the necessity of reaching out for distant resources and the importance of maintaining a collaborative scientific community beyond nationalism during a pandemic.

Viral Dynamics and Immune Correlates of Coronavirus Disease 2019 (COVID-19) Severity.

BACKGROUND: Key knowledge gaps remain in the understanding of viral dynamics and immune response of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) infection. METHODS: We evaluated these characteristics and established their association with clinical severity in a prospective observational cohort study of 100 patients with PCR-confirmed SARS-CoV-2 infection (mean age, 46 years; 56% male; 38% with comorbidities). Respiratory samples (n = 74) were collected for viral culture, serum samples for measurement of IgM/IgG levels (n = 30), and plasma samples for levels of inflammatory cytokines and chemokines (n = 81). Disease severity was correlated with results from viral culture, serologic testing, and immune markers. RESULTS: Fifty-seven (57%) patients developed viral pneumonia, of whom 20 (20%) required supplemental oxygen, including 12 (12%) with invasive mechanical ventilation. Viral culture from respiratory samples was positive for 19 of 74 patients (26%). No virus was isolated when the PCR cycle threshold (Ct) value was >30 or >14 days after symptom onset. Seroconversion occurred at a median (IQR) of 12.5 (9-18) days for IgM and 15.0 (12-20) days for IgG; 54/62 patients (87.1%) sampled at day 14 or later seroconverted. Severe infections were associated with earlier seroconversion and higher peak IgM and IgG levels. Levels of IP-10, HGF, IL-6, MCP-1, MIP-1α, IL-12p70, IL-18, VEGF-A, PDGF-BB, and IL-1RA significantly correlated with disease severity. CONCLUSIONS: We found virus viability was associated with lower PCR Ct value in early illness. A stronger antibody response was associated with disease severity. The overactive proinflammatory immune signatures offer targets for host-directed immunotherapy, which should be evaluated in randomized controlled trials.

Contrastive learning improves critical event prediction in COVID-19 patients.

Deep learning (DL) models typically require large-scale, balanced training data to be robust, generalizable, and effective in the context of healthcare. This has been a major issue for developing DL models for the coronavirus disease 2019 (COVID-19) pandemic, where data are highly class imbalanced. Conventional approaches in DL use cross-entropy loss (CEL), which often suffers from poor margin classification. We show that contrastive loss (CL) improves the performance of CEL, especially in imbalanced electronic health records (EHR) data for COVID-19 analyses. We use a diverse EHR dataset to predict three outcomes: mortality, intubation, and intensive care unit (ICU) transfer in hospitalized COVID-19 patients over multiple time windows. To compare the performance of CEL and CL, models are tested on the full dataset and a restricted dataset. CL models consistently outperform CEL models, with differences ranging from 0.04 to 0.15 for area under the precision and recall curve (AUPRC) and 0.05 to 0.1 for area under the receiver-operating characteristic curve (AUROC).

Differential Cytokine Responses in Hospitalized COVID-19 Patients Limit Efficacy of Remdesivir.

A significant proportion of COVID-19 patients will progress to critical illness requiring invasive mechanical ventilation. This accentuates the need for a therapy that can reduce the severity of COVID-19. Clinical trials have shown the effectiveness of remdesivir in shortening recovery time and decreasing progression to respiratory failure and mechanical ventilation. However, some studies have highlighted its lack of efficacy in patients on high-flow oxygen and mechanical ventilation. This study uncovers some underlying immune response differences between responders and non-responders to remdesivir treatment. Immunological analyses revealed an upregulation of tissue repair factors BDNF, PDGF-BB and PIGF-1, as well as an increase in ratio of Th2-associated cytokine IL-4 to Th1-associated cytokine IFN-γ. Serological profiling of IgG subclasses corroborated this observation, with significantly higher magnitude of increase in Th2-associated IgG2 and IgG4 responses. These findings help to identify the mechanisms of immune regulation accompanying successful remdesivir treatment in severe COVID-19 patients.

Toward a Coronavirus Knowledge Graph.

This study builds a coronavirus knowledge graph (KG) by merging two information sources. The first source is Analytical Graph (AG), which integrates more than 20 different public datasets related to drug discovery. The second source is CORD-19, a collection of published scientific articles related to COVID-19. We combined both chemo genomic entities in AG with entities extracted from CORD-19 to expand knowledge in the COVID-19 domain. Before populating KG with those entities, we perform entity disambiguation on CORD-19 collections using Wikidata. Our newly built KG contains at least 21,700 genes, 2500 diseases, 94,000 phenotypes, and other biological entities (e.g., compound, species, and cell lines). We define 27 relationship types and use them to label each edge in our KG. This research presents two cases to evaluate the KG's usability: analyzing a subgraph (ego-centered network) from the angiotensin-converting enzyme (ACE) and revealing paths between biological entities (hydroxychloroquine and IL-6 receptor; chloroquine and STAT1). The ego-centered network captured information related to COVID-19. We also found significant COVID-19-related information in top-ranked paths with a depth of three based on our path evaluation.

Persistent Symptoms and Association With Inflammatory Cytokine Signatures in Recovered Coronavirus Disease 2019 Patients.

BACKGROUND: The complications and sequelae of coronavirus disease 2019 (COVID-19) and their effect on long-term health are unclear, and the trajectory of associated immune dysregulation is poorly understood. METHODS: We conducted a prospective longitudinal multicenter cohort study at 4 public hospitals in Singapore. Patients with COVID-19 were monitored for a median of 6 months after recovery from acute infection. Clinical symptoms and radiologic data were collected, along with plasma samples for quantification of immune mediators. The relationship between clinical symptoms and immune cytokine profiles was investigated. RESULTS: Two hundred eighty-eight participants were recruited, and follow-up data were available for 183, 175, and 120 participants at days 30, 90, and 180 postsymptom onset, respectively. Symptoms related to COVID-19 were present in 31 (16.9%), 13 (7.4%), and 14 (11.7%) at days 30, 90, and 180. In a multivariable model, age >65 years, non-Chinese ethnicity, and the severity of acute infection were associated with increased likelihood of persistent symptoms. Recovered COVID-19 patients had elevated levels of proinflammatory interleukin (IL)-17A, stem cell factor, IL-12p70, and IL-1ß and pro-angiogenic macrophage inflammatory protein 1ß, brain-derived neurotrophic factor, and vascular endothelial growth factor at day 180 compared with healthy controls. Higher levels of monocyte chemoattractant protein-1 and platelet-derived growth factor-BB were detected in patients with persistent symptoms, versus symptom-free patients. CONCLUSIONS: Approximately 10% of recovered patients had persistent symptoms 6 months after initial infection. Immune cytokine signatures of the recovered patients reflected ongoing chronic inflammation and angiogenesis. Patients with COVID-19 should be monitored closely for emerging long-term health consequences.

Relational Learning Improves Prediction of Mortality in COVID-19 in the Intensive Care Unit.

Traditional Machine Learning (ML) models have had limited success in predicting Coronoavirus-19 (COVID-19) outcomes using Electronic Health Record (EHR) data partially due to not effectively capturing the inter-connectivity patterns between various data modalities. In this work, we propose a novel framework that utilizes relational learning based on a heterogeneous graph model (HGM) for predicting mortality at different time windows in COVID-19 patients within the intensive care unit (ICU). We utilize the EHRs of one of the largest and most diverse patient populations across five hospitals in major health system in New York City. In our model, we use an LSTM for processing time varying patient data and apply our proposed relational learning strategy in the final output layer along with other static features. Here, we replace the traditional softmax layer with a Skip-Gram relational learning strategy to compare the similarity between a patient and outcome embedding representation. We demonstrate that the construction of a HGM can robustly learn the patterns classifying patient representations of outcomes through leveraging patterns within the embeddings of similar patients. Our experimental results show that our relational learning-based HGM model achieves higher area under the receiver operating characteristic curve (auROC) than both comparator models in all prediction time windows, with dramatic improvements to recall.

Analyzing knowledge entities about COVID-19 using entitymetrics.

COVID-19 cases have surpassed the 109 + million markers, with deaths tallying up to 2.4 million. Tens of thousands of papers regarding COVID-19 have been published along with countless bibliometric analyses done on COVID-19 literature. Despite this, none of the analyses have focused on domain entities occurring in scientific publications. However, analysis of these bio-entities and the relations among them, a strategy called entity metrics, could offer more insights into knowledge usage and diffusion in specific cases. Thus, this paper presents an entitymetric analysis on COVID-19 literature. We construct an entity-entity co-occurrence network and employ network indicators to analyze the extracted entities. We find that ACE-2 and C-reactive protein are two very important genes and that lopinavir and ritonavir are two very important chemicals, regardless of the results from either ranking.

Potential effects of SARS-CoV-2 on the gastrointestinal tract and liver.

COVID-19 is a pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Early reported symptoms include fever, cough, and respiratory symptoms. There were few reports of digestive symptoms. However, with COVID-19 spreading worldwide, symptoms such as vomiting, diarrhoea, and abdominal pain have gained increasing attention. Research has found that angiotensin-converting enzyme 2 (ACE2), the SARS-CoV-2 receptor, is strongly expressed in the gastrointestinal tract and liver. Whether theoretically or clinically, many studies have suggested a close connection between COVID-19 and the digestive system. In this review, we summarize the digestive symptoms reported in existing research, discuss the impact of SARS-CoV-2 on the gastrointestinal tract and liver, and determine the possible mechanisms and aetiology, such as cytokine storm. In-depth exploration of the relationship between COVID-19 and the digestive system is urgently needed.