The COVID-19 pandemic represents the most significant public health disaster since the 1918 influenza pandemic. During pandemics such as COVID-19, timely and reliable spatiotemporal forecasting of epidemic dynamics is crucial. Deep learning-based time series models for forecasting have recently gained popularity and have been successfully used for epidemic forecasting. Here we focus on the design and analysis of deep learning-based models for COVID-19 forecasting. We implement multiple recurrent neural network-based deep learning models and combine them using the stacking ensemble technique. In order to incorporate the effects of multiple factors in COVID-19 spread, we consider multiple sources such as COVID-19 confirmed and death case count data and testing data for better predictions. To overcome the sparsity of training data and to address the dynamic correlation of the disease, we propose clustering-based training for high-resolution forecasting.
The methods help us to identify the similar trends of certain groups of regions due to various spatio-temporal effects. We examine the proposed method for forecasting weekly COVID-19 new confirmed cases at county-, state-, and country-level. A comprehensive comparison between different time series models in COVID-19 context is conducted and analyzed. The results show that simple deep learning models can achieve comparable or better performance when compared with more complicated models. We are currently integrating our methods as a part of our weekly forecasts that we provide state and federal authorities.
more »
« less
Informing University COVID-19 Decisions Using Simple Compartmental Models
Tracking the COVID-19 pandemic has been a major challenge for policy makers. Although, several efforts
are ongoing for accurate forecasting of cases, deaths, and hospitalization at various resolutions, few have
been attempted for college campuses despite their potential to become COVID-19 hot-spots. In this paper,
we present a real-time effort towards weekly forecasting of campus-level cases during the fall semester
for four universities in Virginia, United States. We discuss the challenges related to data curation. A
causal model is employed for forecasting with one free time-varying parameter, calibrated against case
data. The model is then run forward in time to obtain multiple forecasts. We retrospectively evaluate
the performance and, while forecast quality suffers during the campus reopening phase, the model makes
reasonable forecasts as the fall semester progresses. We provide sensitivity analysis for the several model
parameters. In addition, the forecasts are provided weekly to various state and local agencies.
more »
« less
- PAR ID:
- 10313656
- Date Published:
- Journal Name:
- ArXivorg
- ISSN:
- 2331-8422
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Larremore, Daniel B (Ed.)more » « less
During the COVID-19 pandemic, forecasting COVID-19 trends to support planning and response was a priority for scientists and decision makers alike. In the United States, COVID-19 forecasting was coordinated by a large group of universities, companies, and government entities led by the Centers for Disease Control and Prevention and the US COVID-19 Forecast Hub (
https://covid19forecasthub.org ). We evaluated approximately 9.7 million forecasts of weekly state-level COVID-19 cases for predictions 1–4 weeks into the future submitted by 24 teams from August 2020 to December 2021. We assessed coverage of central prediction intervals and weighted interval scores (WIS), adjusting for missing forecasts relative to a baseline forecast, and used a Gaussian generalized estimating equation (GEE) model to evaluate differences in skill across epidemic phases that were defined by the effective reproduction number. Overall, we found high variation in skill across individual models, with ensemble-based forecasts outperforming other approaches. Forecast skill relative to the baseline was generally higher for larger jurisdictions (e.g., states compared to counties). Over time, forecasts generally performed worst in periods of rapid changes in reported cases (either in increasing or decreasing epidemic phases) with 95% prediction interval coverage dropping below 50% during the growth phases of the winter 2020, Delta, and Omicron waves. Ideally, case forecasts could serve as a leading indicator of changes in transmission dynamics. However, while most COVID-19 case forecasts outperformed a naïve baseline model, even the most accurate case forecasts were unreliable in key phases. Further research could improve forecasts of leading indicators, like COVID-19 cases, by leveraging additional real-time data, addressing performance across phases, improving the characterization of forecast confidence, and ensuring that forecasts were coherent across spatial scales. In the meantime, it is critical for forecast users to appreciate current limitations and use a broad set of indicators to inform pandemic-related decision making. -
Khudyakov, Yury E (Ed.)We construct an agent-based SEIR model to simulate COVID-19 spread at a 16000-student mostly non-residential urban university during the Fall 2021 Semester. We find that mRNA vaccine coverage at 100% combined with weekly screening testing of 25% of the campus population make it possible to safely reopen to in-person instruction. Our simulations exhibit a right-skew for total infections over the semester that becomes more pronounced with less vaccine coverage, less vaccine effectiveness and no additional preventative measures. This suggests that high levels of infection are not exceedingly rare with campus social connections the main transmission route. Finally, we find that if vaccine coverage is 100% and vaccine effectiveness is above 80%, then a safe reopening is possible even without facemask use. This models possible future scenarios with high coverage of additional “booster” doses of COVID-19 vaccines.more » « less
-
Real-time forecasting of non-stationary time series is a challenging problem, especially when the time series evolves rapidly. For such cases, it has been observed that ensemble models consisting of a diverse set of model classes can perform consistently better than individual models. In order to account for the nonstationarity of the data and the lack of availability of training examples, the models are retrained in real-time using the most recent observed data samples. Motivated by the robust performance properties of ensemble models, we developed a Bayesian model averaging ensemble technique consisting of statistical, deep learning, and compartmental models for fore-casting epidemiological signals, specifically, COVID-19 signals. We observed the epidemic dynamics go through several phases (waves). In our ensemble model, we observed that different model classes performed differently during the various phases. Armed with this understanding, in this paper, we propose a modification to the ensembling method to employ this phase information and use different weighting schemes for each phase to produce improved forecasts. However, predicting the phases of such time series is a significant challenge, especially when behavioral and immunological adaptations govern the evolution of the time series. We explore multiple datasets that can serve as leading indicators of trend changes and employ transfer entropy techniques to capture the relevant indicator. We propose a phase prediction algorithm to estimate the phases using the leading indicators. Using the knowledge of the estimated phase, we selectively sample the training data from similar phases. We evaluate our proposed methodology on our currently deployed COVID-19 forecasting model and the COVID-19 ForecastHub models. The overall performance of the proposed model is consistent across the pandemic. More importantly, it is ranked second during two critical rapid growth phases in cases, regimes where the performance of most models from the ForecastHub dropped significantly.more » « less
-
When college campuses resumed in-person learning opportunities following initial lockdowns during the COVID-19 pandemic, many facets of campus life looked different. These differences continue to evolve from semester to semester because of changing health guidelines, school decisions, and personal convictions. Academic makerspaces were not exempt from these changes and have experienced fluctuating usage and usage barriers over the past several semesters. Better understanding the effects of COVID-19 on academic makerspaces can help ensure that students continue to draw maximum benefits from these learning spaces and also provides potential advice for administrators and educators for future disturbances. Data collected via tool usage surveys administered to makerspace users at a large public university during the three semesters following the start of the pandemic (Fall 2020, Spring 2021, and Spring 2022) is used here to investigate. COVID-19 restrictions present during Fall 2020 and Spring 2021 were mostly loosened in Spring 2022. The makerspace is modeled as a bipartite network, with student and tool interactions determined via end-of-semester surveys. The network is analyzed using nestedness, a metric primarily used in ecology to evaluate the stability of an ecosystem and proposed here as a quantitative method to evaluate makerspace health. The surveys used to create the network models also provide validation, as students were asked to share tools used during the semester in question. The results suggest that nestedness is linearly proportional to usage, both increases and decreases. As such, tracking the nestedness of a makespace over time can serve as a warning that unintended restrictions are in place, intentional restrictions and/or policies may be too severe, or whether a space has effectively recovered from temporary restrictions.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1101/2021.07.01.21259851
