• Publicationsclose
• Research softwareclose
• Other research productsclose
• 2018-2022close
• Preprint close
• English close
• COVID-19 close
• Digital Humanities and Cultural Heritage close

Abstract. The world experienced the beginning of the COVID-19 pandemic by the end of 2019 to the beginning of 2020. Governments implemented strategies to contain it, most based on lockdowns. Mexico was no exception. The lockdown was initiated in March 2020, and with it, a reduction in the seismic noise level was witnessed by the seismic stations of the national and Valley of Mexico networks. Stations located in municipalities with more than 50 000 people usually experience larger seismic noise levels at frequencies between 1 and 5 Hz, associated with human activity. The largest noise levels are recorded in Mexico City, which has the largest population in the country. The largest drop was observed in Hermosillo, Sonora; however, it was also the city with the fastest return to activities, which seems to correlate with a quick increase in confirmed COVID-19 cases. Mexico initiated a traffic-light system to modulate the re-opening of economic activities for each state. Therefore, since 1 June, noise levels have generally reflected the colour of the state traffic light. Furthermore, the reduction in the noise level at seismic stations has allowed identification of smaller earthquakes without signal processing. Also, people in cities have perceived smaller or more distant quakes.

The massive spread of false information on social media has become a global risk especially in a global pandemic situation like COVID-19. False information detection has thus become a surging research topic in recent months. NLP4IF-2021 shared task on fighting the COVID-19 infodemic has been organised to strengthen the research in false information detection where the participants are asked to predict seven different binary labels regarding false information in a tweet. The shared task has been organised in three languages; Arabic, Bulgarian and English. In this paper, we present our approach to tackle the task objective using transformers. Overall, our approach achieves a 0.707 mean F1 score in Arabic, 0.578 mean F1 score in Bulgarian and 0.864 mean F1 score in English ranking 4th place in all the languages. Accepted to Workshop on NLP for Internet Freedom (NLP4IF) at NAACL 2021

AbstractMuch research has shown that the first lockdowns imposed in response to the COVID-19 pandemic were associated with changes in routine activities and, therefore, changes in crime. While several types of violent and property crime decreased immediately after the first lockdown, online crime rates increased. Nevertheless, little research has explored the relationship between multiple lockdowns and crime in the mid-term. Furthermore, few studies have analysed potentially contrasting trends in offline and online crimes using the same dataset. To fill these gaps in research, the present article employs interrupted time-series analysis to examine the effects on offline and online crime of the three lockdown orders implemented in Northern Ireland. We analyse crime data recorded by the police between April 2015 and May 2021. Results show that many types of traditional offline crime decreased after the lockdowns but that they subsequently bounced back to pre-pandemic levels. In contrast, results appear to indicate that cyber-enabled fraud and cyber-dependent crime rose alongside lockdown-induced changes in online habits and remained higher than before COVID-19. It is likely that the pandemic accelerated the long-term upward trend in online crime. We also find that lockdowns with stay-at-home orders had a clearer impact on crime than those without. Our results contribute to understanding how responses to pandemics can influence crime trends in the mid-term as well as helping identify the potential long-term effects of the pandemic on crime, which can strengthen the evidence base for policy and practice.

Several studies have been devoted to establishing the effects of the COVID-19 pandemic on mental health across gender, age, and ethnicity. However, much less attention has been paid to the differential effect of COVID-19 according to different personalities. We do this using the UK Household Longitudinal Study (UKHLS), a large-scale panel survey representative of the UK population. The UKHLS allows us to assess the mental health of the same respondent before and during the COVID-19 period based on their “Big Five” personality traits and cognitive skills. We find that during the COVID-19 period, individuals who have more extravert and open personality traits report a higher mental health deterioration, while those scoring higher in agreeableness are less affected. The effect of openness is particularly strong: One more SD predicts up to 0.23 more symptoms of mental health deterioration in the 12-item General Health Questionnaire (GHQ-12) test during the COVID-19 period. In particular, for females, cognitive skills and openness are strong predictors of mental health deterioration, while for non-British White respondents, these predictors are extraversion and openness. Neuroticism strongly predicts worse mental health cross-sectionally, but it does not lead to significantly stronger deterioration during the pandemic. The study’s results are robust to the inclusion of potential confounding variables such as changes in physical health, household income, and job status (like unemployed or furloughed).

IntroductionThis study presents COVID-Twitter-BERT (CT-BERT), a transformer-based model that is pre-trained on a large corpus of COVID-19 related Twitter messages. CT-BERT is specifically designed to be used on COVID-19 content, particularly from social media, and can be utilized for various natural language processing tasks such as classification, question-answering, and chatbots. This paper aims to evaluate the performance of CT-BERT on different classification datasets and compare it with BERT-LARGE, its base model.MethodsThe study utilizes CT-BERT, which is pre-trained on a large corpus of COVID-19 related Twitter messages. The authors evaluated the performance of CT-BERT on five different classification datasets, including one in the target domain. The model's performance is compared to its base model, BERT-LARGE, to measure the marginal improvement. The authors also provide detailed information on the training process and the technical specifications of the model.ResultsThe results indicate that CT-BERT outperforms BERT-LARGE with a marginal improvement of 10-30% on all five classification datasets. The largest improvements are observed in the target domain. The authors provide detailed performance metrics and discuss the significance of these results.DiscussionThe study demonstrates the potential of pre-trained transformer models, such as CT-BERT, for COVID-19 related natural language processing tasks. The results indicate that CT-BERT can improve the classification performance on COVID-19 related content, especially on social media. These findings have important implications for various applications, such as monitoring public sentiment and developing chatbots to provide COVID-19 related information. The study also highlights the importance of using domain-specific pre-trained models for specific natural language processing tasks. Overall, this work provides a valuable contribution to the development of COVID-19 related NLP models.

Deep neural networks have increasingly been used as an auxiliary tool in healthcare applications, due to their ability to improve performance of several diagnosis tasks. However, these methods are not widely adopted in clinical settings due to the practical limitations in the reliability, generalizability, and interpretability of deep learning based systems. As a result, methods have been developed that impose additional constraints during network training to gain more control as well as improve interpretabilty, facilitating their acceptance in healthcare community. In this work, we investigate the benefit of using Orthogonal Spheres (OS) constraint for classification of COVID-19 cases from chest X-ray images. The OS constraint can be written as a simple orthonormality term which is used in conjunction with the standard cross-entropy loss during classification network training. Previous studies have demonstrated significant benefits in applying such constraints to deep learning models. Our findings corroborate these observations, indicating that the orthonormality loss function effectively produces improved semantic localization via GradCAM visualizations, enhanced classification performance, and reduced model calibration error. Our approach achieves an improvement in accuracy of 1.6% and 4.8% for two- and three-class classification, respectively; similar results are found for models with data augmentation applied. In addition to these findings, our work also presents a new application of the OS regularizer in healthcare, increasing the post-hoc interpretability and performance of deep learning models for COVID-19 classification to facilitate adoption of these methods in clinical settings. We also identify the limitations of our strategy that can be explored for further research in future. Accepted in the 2021 ACM CHIL Workshop track. An extended version of this work is under consideration at Pattern Recognition Letters

SUMMARYRecent surveillance has revealed the emergence of the SARS-CoV-2 Omicron variant (BA.1/B.1.1.529) harboring up to 36 mutations in spike protein, the target of vaccine-induced neutralizing antibodies. Given its potential to escape vaccine-induced humoral immunity, we measured neutralization potency of sera from 88 mRNA-1273, 111 BNT162b, and 40 Ad26.COV2.S vaccine recipients against wild type, Delta, and Omicron SARS-CoV-2 pseudoviruses. We included individuals that were vaccinated recently (<3 months), distantly (6-12 months), or recently boosted, and accounted for prior SARS-CoV-2 infection. Remarkably, neutralization of Omicron was undetectable in most vaccinated individuals. However, individuals boosted with mRNA vaccines exhibited potent neutralization of Omicron only 4-6-fold lower than wild type, suggesting that boosters enhance the cross-reactivity of neutralizing antibody responses. In addition, we find Omicron pseudovirus is more infectious than any other variant tested. Overall, this study highlights the importance of boosters to broaden neutralizing antibody responses against highly divergent SARS-CoV-2 variants.

Abstract Researchers around the globe are searching for a "combo-drug" against Covid-19 by trying to combine various existing drugs. Given a set of such drugs, various algorithms (based, for example, on artificial intelligence) are used to identify the efficacy of different shares of the constituent drugs in the combo-drug. Namely, the relative weight of each drug in a "cooperative" scheme of therapy is sought-after. In the current note we propose to identify these weights using the theory of cooperative games, and in particular the Shapley value, one of the fundamental solution concepts of such games. We derive the weight of each drug by its (normalized) average marginal contribution over all possible "coalitions" of drugs it is used with, where a drug's marginal contribution to a coalition is defined as the increase in the coalition's probability to act against a virus should the drug become its "member". Hence we endow each drug with a consistent measure of significance (which is due to the consistency that Shapley value is associated with). At a theoretical level, we build the cooperative game, and compute the Shapley values, within a milestone model in drug combination theory, the Bliss independence model. At a practical level, the predictions of our game-theoretic model can be tested by using in-vitro experiments, namely experiments that are conducted in test tubes.

We build a deep-learning-based SEIR-AIM model integrating the classical Susceptible-Exposed-Infectious-Removed epidemiology model with forecast modules of infection, community mobility, and unemployment. Through linking Google's multi-dimensional mobility index to economic activities, public health status, and mitigation policies, our AI-assisted model captures the populace's endogenous response to economic incentives and health risks. In addition to being an effective predictive tool, our analyses reveal that the long-term effective reproduction number of COVID-19 equilibrates around one before mass vaccination using data from the United States. We identify a "policy frontier" and identify reopening schools and workplaces to be the most effective. We also quantify protestors' employment-value-equivalence of the Black Lives Matter movement and find that its public health impact to be negligible. Preprint, not peer reviewed

Background: There remain many unknowns regarding the natural history, onset, distribution and both the individual and population burden of the ongoing COVID-19 pandemic associated with the spread of the SARS-CoV-2 virus. Here, we used a combination of classic epidemiological methods, natural language processing (NLP), and machine learning (for predictive modelling), to analyse the clinical information in the electronic health records (EHRs) of patients with COVID-19. This approach holds the potential to better define the disease and its associated outcomes, most notably ICU admission. Methods: This is a multicentre, non-interventional, retrospective study using the unstructured free-text clinical information captured in the EHRs of the participating hospital sites within the SESCAM Healthcare Network (Castilla La-Mancha, Spain, with 2.035 M inhabitants). We collected clinical information from the entire population with available EHRs (1,364,924 patients) for the period comprised between January 1, 2020 and March 29, 2020. Following identification of all COVID-19 cases seen in hospitals and primary care settings (all departments), we extracted related information upon diagnosis (including demographic characteristics, symptoms upon diagnosis, and other clinical information) and during disease progression and outcome (admission, discharge, and ICU admission). A data-driven analysis explored the minimum set of clinical variables associated with requiring ICU admission. Findings: A total of 10,504 patients with a clinical or PCR-confirmed diagnosis of COVID-19 were identified, 52.5% males, with a mean±SD age of 58.2±19.7 years, and age distribution ranging from 39oC/102oF without respiratory crackles), were free of ICU admission. On the contrary, COVID-19 patients aged 40 to 79 years were likely to be admitted to the ICU if they had tachypnoea and delayed their visit to the ER after being seen in primary care. Interpretation: Our results show that a combination of easily obtained clinical variables (age, fever, and tachypnoea with/without respiratory crackles) predicts which COVID-19 patients require ICU admission. Funding Statement: This study was sponsored by SAVANA (https://www.savanamed.com/) Declaration of Interests: None to declare Ethics Approval Statement: This study was classified as a ‘non-post-authorization study’ (EPA) by the Spanish Agency of Medicines and Health Products (AEMPS), and it was approved by the Research Ethics Committee at the University Hospital of Guadalajara (Spain).