Forecasting Emerging Pandemics with Transfer Learning and Location-aware News Analysis

Monitoring and forecasting epidemic diseases are of prime importance to public health organizations and policymakers in taking proper measures and adjusting prevention tactics. Early prediction is especially important to restrict the spread of emerging pandemics such as COVID-19. However, despite increasing research and development for various epidemics, several challenges remain unresolved. On the one hand, early-stage epidemic prediction for emerging new diseases is difficult because of data paucity and lack of experience. On the other hand, many existing studies ignore or fail to leverage the contribution of social factors such as news, geolocations, and climate. Even though some researchers have recognized the profound impact of social features, capturing the dynamic correlation between these features and pandemics requires an extensive understanding of heterogeneous formats of data and mechanisms. In this paper, we design TLSS, a neural transfer learning architecture for learning and transferring general characteristics of existing epidemic diseases to predict a new pandemic. We propose a new feature module to learn the impact of news sentiment and semantic information on epidemic transmission. We then combine this information with historical time-series features to forecast future infection cases in a dynamic propagation process. We compare the proposed model with several state-of-the-art statistics approaches and deep learning methods in epidemic prediction with different lead times of ground truth. We conducted extensive experiments on three stages of COVID-19 development in the United States. Our experiment demonstrates that our approach has strong predictive performance for COVID infection cases, especially with longer lead times. © 2022 IEEE.

[Study of incubation period of infection with 2019-nCoV Omicron variant BA.5.1.3].

Objective: To study the incubation period of the infection with 2019-nCoV Omicron variant BA.5.1.3. Methods: Based on the epidemiological survey data of 315 COVID-19 cases and the characteristics of interval censored data structure, log-normal distribution and Gamma distribution were used to estimate the incubation. Bayes estimation was performed for the parameters of each distribution function using discrete time Markov chain Monte Carlo algorithm. Results: The mean age of the 315 COVID-19 cases was (42.01±16.54) years, and men accounted for 30.16%. A total of 156 cases with mean age of (41.65±16.32) years reported the times when symptoms occurred. The log-normal distribution and Gamma distribution indicated that the M (Q1, Q3) of the incubation period from exposure to symptom onset was 2.53 (1.86, 3.44) days and 2.64 (1.91, 3.52) days, respectively, and the M (Q1, Q3) of the incubation period from exposure to the first positive nucleic acid detection was 2.45 (1.76, 3.40) days and 2.57 (1.81, 3.52) days, respectively. Conclusions: The incubation period by Bayes estimation based on log-normal distribution and Gamma distribution, respectively, was similar to each other, and the best distribution of incubation period was Gamma distribution, the difference between the incubation period from exposure to the first positive nucleic acid detection and the incubation period from exposure to symptom onset was small. The median of incubation period of infection caused by Omicron variant BA.5.1.3 was shorter than those of previous Omicron variants.

COVID-19 vaccine acceptance and coverage among pregnant persons in the United States.

Pregnant persons are at higher risk of severe COVID-19. Although vaccination is recommended, COVID-19 vaccination rates are lower among pregnant persons compared to the non-pregnant population. We aimed to evaluate acceptance of any dose of COVID-19 vaccine during pregnancy. A national online cross-sectional survey of US adults who were pregnant between December 2020 and July 2021 was used to measure COVID-19 vaccine behaviors, attitudes, and beliefs. Post-stratification weighting was used to ensure representativeness to the US population. Marginal log-binomial models were used to estimate adjusted prevalence ratios (aPR) of COVID-19 vaccine acceptance, accounting for sociodemographic factors. Of 5,660 who responded to survey advertisements, 2,213 met eligibility criteria and completed the survey; 55.4% of respondents received or planned to receive COVID-19 vaccine prior to or during pregnancy, 27.0% planned to vaccinate after pregnancy, 8.8% were unsure and 8.7% had no plans to vaccinate. Individuals were more likely to receive or plan to receive COVID-19 vaccine if they had group prenatal care (aPR 1.57; 95% CI 1.40, 1.75), were employed in a workplace with a policy recommending vaccination (aPR 1.15; 95% CI 1.06, 1.26), and believed COVID-19 vaccines are safe (aPR 2.86; 95% CI 2.49, 3.29). Pregnant persons who were recommended COVID-19 vaccination by their healthcare provider less commonly reported concerns about vaccine safety (35.5% vs 55.9%) and were more likely to accept COVID-19 vaccines (aPR 1.52; 95% CI 1.31, 1.76). COVID-19 vaccine acceptance during pregnancy is not universal and public health intervention will be needed to continue to increase vaccine coverage.

Recurrent Multi-task Graph Convolutional Networks for COVID-19 Knowledge Graph Link Prediction

Knowledge graphs (KGs) are a way to model data involving intricate relations between a number of entities. Understanding the information contained in KGs and predicting what hidden relations may be present can provide valuable domain-specific knowledge. Thus, we use data provided by the 5th Annual Oak Ridge National Laboratory Smoky Mountains Computational Sciences Data Challenge 2 as well as auxiliary textual data processed with natural language processing techniques to form and analyze a COVID-19 KG of biomedical concepts and research papers. Moreover, we propose a recurrent graph convolutional network model that predicts both the existence of novel links between concepts in this COVID-19 KG and the time at which the link will form. We demonstrate our model’s promising performance against several baseline models. The utilization of our work can give insights that are useful in COVID-19-related fields such as drug development and public health. All code for our paper is publicly available at https://github.com/RemingtonKim/SMCDC2021. © 2022, Springer Nature Switzerland AG.