Background To accompany the lifting of COVID-19 lockdown measures, Luxembourg implemented a mass screening (MS) programme. The first phase coincided with an early summer epidemic wave in 2020. Methods rRT-PCR-based screening for SARS-CoV-2 was performed by pooling of samples. The infrastructure allowed the testing of the entire resident and cross-border worker populations. The strategy relied on social connectivity within different activity sectors. Invitation frequencies were tactically increased in sectors and regions with higher prevalence. The results were analysed alongside contact tracing data. Findings The voluntary programme covered 49% of the resident and 22% of the cross-border worker populations. It identified 850 index cases with an additional 249 cases from contact tracing. Over-representation was observed in the services, hospitality and construction sectors alongside regional differences. Asymptomatic cases had a significant but lower secondary attack rate when compared to symptomatic individuals. Based on simulations using an agent-based SEIR model, the total number of expected cases would have been 42·9% (90% CI [-0·3, 96·7]) higher without MS. Mandatory participation would have resulted in a further difference of 39·7% [19·6, 59·2]. Interpretation Strategic and tactical MS allows the suppression of epidemic dynamics. Asymptomatic carriers represent a significant risk for transmission. Containment of future outbreaks will depend on early testing in sectors and regions. Higher participation rates must be assured through targeted incentivisation and recurrent invitation. Funding This project was funded by the Luxembourg Ministries of Higher Education and Research, and Health.

The rapid spread of the Coronavirus SARS-2 is a major challenge that led almost all governments worldwide to take drastic measures to respond to the tragedy. Chief among those measures is the massive lockdown of entire countries and cities, which beyond its global economic impact has created some deep social and psychological tensions within populations. While the adopted mitigation measures (including the lockdown) have generally proven useful, policymakers are now facing a critical question: how and when to lift the mitigation measures? A carefully-planned exit strategy is indeed necessary to recover from the pandemic without risking a new outbreak. Classically, exit strategies rely on mathematical modeling to predict the effect of public health interventions. Such models are unfortunately known to be sensitive to some key parameters, which are usually set based on rules-of-thumb. In this paper, we propose to augment epidemiological forecasting with actual data-driven models that will learn to fine-tune predictions for different contexts (e.g., per country). We have therefore built a pandemic simulation and forecasting toolkit that combines a deep learning estimation of the epidemiological parameters of the disease in order to predict the cases and deaths, and a genetic algorithm component searching for optimal trade-offs/policies between constraints and objectives set by decision-makers. Replaying pandemic evolution in various countries, we experimentally show that our approach yields predictions with much lower error rates than pure epidemiological models in 75% of the cases and achieves a 95% R2 score when the learning is transferred and tested on unseen countries. When used for forecasting, this approach provides actionable insights into the impact of individual measures and strategies.

Background Following a first wave in spring and gradual easing of lockdown, Luxembourg experienced an early second epidemic wave of SARS-CoV-2 before the start of summer school holidays on 15th July. This provided the opportunity to investigate the role of school-age children and school settings for transmission. Methods We compared the incidence of SARS-CoV-2 in school-age children, teachers and the general working population in Luxembourg during two epidemic waves: a spring wave from March–April 2020 corresponding to general lockdown with schools being closed and May–July 2020 corresponding to schools being open. We assessed the number of secondary transmissions occurring in schools between May and July 2020 using routine contact tracing data. Results During the first wave in March–April 2020 when schools were closed, the incidence in pupils peaked at 28 per 100,000, while during the second wave in May–July 2020 when schools were open, incidence peaked 100 per 100,000. While incidence of SARS-CoV-2 was higher in adults than in children during the first spring wave, no significant difference was observed during the second wave in early summer. Between May and July 2020, we identified a total of 390 and 34 confirmed COVID-19 cases among 90,150 school-age children and 11,667 teachers, respectively. We further estimate that 179 primary cases caused 49 secondary cases in schools. While some small clusters of mainly student-to-student transmission within the same class were identified, we did not observe any large outbreaks with multiple generations of infection. Conclusions Transmission of SARS-CoV-2 within Luxembourg schools was limited during an early summer epidemic wave in 2020. Precautionary measures including physical distancing as well as easy access to testing, systematic contact tracing appears to have been successful in mitigating transmission within educational settings.

No abstract

No abstract

Named Entity Recognition (NER) is a fundamental Natural Language Processing (NLP) task and has remained an active research field. In recent years, transformer models and more specifically the BERT model developed at Google revolutionised the field of NLP. While the performance of transformer-based approaches such as BERT has been studied for NER, there has not yet been a study for the fine-grained Named Entity Recognition (FG-NER) task. In this paper, we compare three transformer-based models (BERT, RoBERTa, and XLNet) to two non-transformer-based models (CRF and BiLSTM-CNN-CRF). Furthermore, we apply each model to a multitude of distinct domains. We find that transformer-based models incrementally outperform the studied non-transformer-based models in most domains with respect to the F1 score. Furthermore, we find that the choice of domain significantly influenced the performance regardless of the respective data size or the model chosen.

Neural networks for language modeling have been proven effective on several sub-tasks of natural language processing. Training deep language models, however, is time-consuming and computationally intensive. Pre-trained language models such as BERT are thus appealing since (1) they yielded state-of-the-art performance, and (2) they offload practitioners from the burden of preparing the adequate resources (time, hardware, and data) to train models. Nevertheless, because pre-trained models are generic, they may underperform on specific domains. In this study, we investigate the case of multi-class text classification, a task that is relatively less studied in the literature evaluating pre-trained language models. Our work is further placed under the industrial settings of the financial domain. We thus leverage generic benchmark datasets from the literature and two proprietary datasets from our partners in the financial technological industry. After highlighting a challenge for generic pre-trained models (BERT, DistilBERT, RoBERTa, XLNet, XLM) to classify a portion of the financial document dataset, we investigate the intuition that a specialized pre-trained model for financial documents, such as FinBERT, should be leveraged. Nevertheless, our experiments show that the FinBERT model, even with an adapted vocabulary, does not lead to improvements compared to the generic BERT models. CCS CONCEPTS• Applied computing → Text processing.

No abstract