Search

Accurate forecasting and analysis of emerging pandemics play a crucial role in effective public health management and decision-making. Traditional approaches primarily rely on epidemiological data, overlooking other valuable sources of information that could act as sensors or indicators of pandemic patterns. In this paper, we propose a novel framework, MGLEP, that integrates temporal graph neural networks and multi-modal data for learning and forecasting. We incorporate big data sources, including social media content, by utilizing specific pre-trained language models and discovering the underlying graph structure among users. This integration provides rich indicators of pandemic dynamics through learning with temporal graph neural networks. Extensive experiments demonstrate the effectiveness of our framework in pandemic forecasting and analysis, outperforming baseline methods across different areas, pandemic situations, and prediction horizons. The fusion of temporal graph learning and multi-modal data enables a comprehensive understanding of the pandemic landscape with less time lag, cheap cost, and more potential information indicators., (© 2024. The Author(s).)

A situation of emergency involving the whole population introduces changes in the dynamics of the health services that are provided. The magnitude of these shifts should be also linked to the medical speciality within which the health benefits are delivered. The aim of the paper was to identify changes in tendencies in inpatient medical service delivery during the COVID-19 pandemic by medical specialties. On the basis of a database extracted for in-patient treatment received from the public payer, a retrospective analysis was carried out. Comparing the values of the dynamics of the services provided in each medical speciality, the period before the COVID-19 pandemic was collated to the years of the pandemic (2020-2021). In the period before COVID-19, positive patient dynamics were observed in more than half of the specialities. Between 2020 and 2021, virtually all specialties reversed the trend and negative dynamics were recorded. The dynamics in 2021 indicate a process of return to the values from 2015 to 2019. Emergency situation has affected the dynamics of healthcare provision in different specialities to various extent. The most resistant to the negative impulses of the state of emergency were the areas that are strictly organisationally and financially defined (e.g. the group of "therapeutic and drug programmes")., (© 2024. The Author(s).)

In this paper, NeuralProphet (NP), an explainable hybrid modular framework, enhances the forecasting performance of pandemics by adding two neural network modules; auto-regressor (AR) and lagged-regressor (LR). An advanced deep auto-regressor neural network (Deep-AR-Net) model is employed to implement these two modules. The enhanced NP is optimized via AdamW and Huber loss function to perform multivariate multi-step forecasting contrast to Prophet. The models are validated with COVID-19 time-series datasets. The NP's efficiency is studied component-wise for a long-term forecast for India and an overall reduction of 60.36% and individually 34.7% by AR-module, 53.4% by LR-module in MASE compared to Prophet. The Deep-AR-Net model reduces the forecasting error of NP for all five countries, on average, by 49.21% and 46.07% for short-and-long-term, respectively. The visualizations confirm that forecasting curves are closer to the actual cases but significantly different from Prophet. Hence, it can develop a real-time decision-making system for highly infectious diseases., (© 2024. The Author(s).)

A global disaster, such as the recent Covid-19 pandemic, affects every aspect of our lives and there is a need to investigate these highly complex phenomena if one aims to diminish their impact in the health of the population, as well as their socio-economic stability. In this paper we present an attempt to understand the role of the governmental authorities and the response of the rest of the population facing such emergencies. We present a mathematical model that takes into account the epidemiological features of the pandemic and also the actions of people responding to it, focusing only on three aspects of the system, namely, the fear of catching this serious disease, the impact on the economic activities and the compliance of the people to the mitigating measures adopted by the authorities. We apply the model to the specific case of Spain, since there are accurate data available about these three features. We focused on tourism as an example of the economic activity, since this sector of economy is one of the most likely to be affected by the restrictions imposed by the authorities, and because it represents an important part of Spanish economy. The results of numerical calculations agree with the empirical data in such a way that we can acquire a better insight of the different processes at play in such a complex situation, and also in other different circumstances. [ABSTRACT FROM AUTHOR]

The emergence of long COVID during the ongoing COVID-19 pandemic has presented considerable challenges for healthcare professionals and researchers. The task of identifying relevant literature is particularly daunting due to the rapidly evolving scientific landscape, inconsistent definitions, and a lack of standardized nomenclature. This paper proposes a novel solution to this challenge by employing machine learning techniques to classify long COVID literature. However, the scarcity of annotated data for machine learning poses a significant obstacle. To overcome this, we introduce a strategy called medical paraphrasing, which diversifies the training data while maintaining the original content. Additionally, we propose a Data-Reweighting-Based Multi-Level Optimization Framework for Domain Adaptive Paraphrasing, supported by a Meta-Weight-Network (MWN). This innovative approach incorporates feedback from the downstream text classification model to influence the training of the paraphrasing model. During the training process, the framework assigns higher weights to the training examples that contribute more effectively to the downstream task of long COVID text classification. Our findings demonstrate that this method substantially improves the accuracy and efficiency of long COVID literature classification, offering a valuable tool for physicians and researchers navigating this complex and ever-evolving field.

This study examined the impact of the COVID-19 pandemic on marriage, divorce, birth, and death rates using the Poisson regression model and an interrupted time-series Poisson regression model. Before the pandemic, marriage and birth rates were decreasing, while divorce and death rates were increasing, with only the trend in birth rates being statistically significant. The immediate effect of the pandemic was a significant decrease in the divorce rate, but there were non-significant effects on birth and marriage rates. However, in the months following the onset of the pandemic, there was a statistically significant sustained effect on increasing death and divorce rates. Forecasts based on pre-pandemic data showed that by the end of 2020, marriage, divorce, death, and birth rates were higher compared to pre-pandemic levels. In conclusion, the pandemic has greatly impacted society, particularly in terms of death and divorce rates. Birth rates were not immediately affected to the time lag between decisions and actual births. Fear of COVID-19 may have increased death rates as people avoided seeking medical help. Vaccination and effective treatment strategies are vital in reducing the pandemic's impact on mortality. Supporting families financially is important due to the role of economic issues in couples' decisions. [ABSTRACT FROM AUTHOR]

How does threat from disease shape our cooperative actions and the social norms that guide such behaviour? To study these questions, we draw on a collective-risk social dilemma experiment that we ran before the emergence of the Covid-19 pandemic (Wave 1, 2018) and compare this to its exact replication, sampling from the same population, that we conducted during the first wave of the pandemic (Wave 2, 2020). Tightness-looseness theory predicts and evidence generally supports that both cooperation and accompanying social norms should increase, yet, we mostly did not find this. Contributions, the probability of reaching the threshold (cooperation), and the contents of the social norm (how much people should contribute) remained similar across the waves, although the strength of these social norms were slightly greater in Wave 2. We also study whether the results from Wave 1 that should not be affected by the pandemic—the relationship between social norms and cooperation and specific behavioural types—replicate in Wave 2 and find that these results generally hold. Overall, our work demonstrates that social norms are important drivers of cooperation, yet, communicable diseases, at least in the short term, have little or no effects on either. [ABSTRACT FROM AUTHOR]

The persistent SARS-CoV-2 pandemic, spanning over three years, has profoundly impacted daily life worldwide. Crucial measures like interpersonal distance (IPD) and mask-wearing have become paramount in preventing infection. With SARS-CoV-2 now resembling an endemic condition similar to influenza, it is vital to assess the changes in IPD influenced by relevant factors during and after the pandemic. This study concentrated on two specific stages (the pandemic stage and the post-pandemic era) and investigated variations in IPD with different test combinations. Variables taken into account encompassed the pandemic stage, participant gender, target gender, and mask-wearing status. We examined IPD data from 100 young individuals (50 males and 50 females) at each stage, with a one-year interval between tests. The results highlighted the substantial impact of all variables on perceived IPD during the pandemic phase (all p < 0.001). However, in the post-pandemic stage, only mask-wearing demonstrated a notable effect on IPD (p < 0.001). As the SARS-CoV-2 epidemic subsides, the enduring influence of mask usage on IPD persists. Nevertheless, the gap between the two mask-wearing scenarios diminishes, contracting from around 50 cm during the epidemic phase to 20 cm in the post-epidemic phase. Across these two pandemic stages, there was an overall reduction of approximately 90 cm in IPD, indicating a noteworthy decrease in perceived personal space and a consequential shortening of social proximity during the post-pandemic stage. This decrease in IPD may suggest the successful socio-cultural adaptation of the young Taiwanese individuals in our study during the post-pandemic era. [ABSTRACT FROM AUTHOR]