DeepCOVID: An Operational Deep Learning-driven Framework for Explainable Real-time COVID-19 Forecasting

How do we forecast an emerging pandemic in real time in a purely data-driven manner? How to leverage rich heterogeneous data based on various signals such as mobility, testing, and/or disease exposure for forecasting? How to handle noisy data and generate uncertainties in the forecast? In this paper, we present DeepCOVID, an operational deep learning framework designed for real-time COVID-19 forecasting. DeepCOVID works well with sparse data and can handle noisy heterogeneous data signals by propagating the uncertainty from the data in a principled manner resulting in meaningful uncertainties in the forecast. The framework also consists of modules for both real-time and retrospective exploratory analysis to enable interpretation of the forecasts. Results from real-time predictions (featured on the CDC website and FiveThirtyEight.com) since April 2020 indicates that our approach is competitive among the methods in the COVID-19 Forecast Hub, especially for short-term predictions.

[1]  E. Nsoesie,et al.  Monitoring Influenza Epidemics in China with Search Query from Baidu , 2013, PloS one.

[2]  Ronald Rosenfeld,et al.  Flexible Modeling of Epidemics with an Empirical Bayes Framework , 2014, PLoS Comput. Biol..

[3]  Y. Wang,et al.  Ensemble Forecasts of Coronavirus Disease 2019 (COVID-19) in the U.S. , 2020, medRxiv.

[4]  Vivek F. Farias,et al.  The Limits to Learning an SIR Process: Granular Forecasting for Covid-19 , 2020, ArXiv.

[5]  Johannes Bracher,et al.  Evaluating epidemic forecasts in an interval format , 2020, PLoS Comput. Biol..

[6]  C. Buckee,et al.  Wrong but Useful - What Covid-19 Epidemiologic Models Can and Cannot Tell Us. , 2020, The New England journal of medicine.

[7]  Aaron C. Miller,et al.  A Smartphone-Driven Thermometer Application for Real-time Population- and Individual-Level Influenza Surveillance , 2018, Clinical infectious diseases : an official publication of the Infectious Diseases Society of America.

[8]  J. Shaman,et al.  Forecasting seasonal outbreaks of influenza , 2012, Proceedings of the National Academy of Sciences.

[9]  Ellyn Ayton,et al.  Forecasting influenza-like illness dynamics for military populations using neural networks and social media , 2017, PloS one.

[10]  Jessica T Davis,et al.  The effect of travel restrictions on the spread of the 2019 novel coronavirus (COVID-19) outbreak , 2020, Science.

[11]  Reid Priedhorsky,et al.  Dynamic Bayesian Influenza Forecasting in the United States with Hierarchical Discrepancy (with Discussion) , 2017, Bayesian Analysis.

[12]  Sergey Ioffe,et al.  Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift , 2015, ICML.

[13]  J. Hageman.  The Coronavirus Disease 2019 (COVID-19). , 2020, Pediatric annals.

[14]  Alessandro Vespignani,et al.  Real-time numerical forecast of global epidemic spreading: case study of 2009 A/H1N1pdm , 2012, BMC Medicine.

[15]  Kevin R. Williams,et al.  Measuring Movement and Social Contact with Smartphone Data: A Real-Time Application to Covid-19 , 2020, SSRN Electronic Journal.

[16]  Alfonso J. Rodriguez-Morales,et al.  Coronavirus Disease 2019–COVID-19 , 2020, Clinical Microbiology Reviews.

[17]  R. Rosenfeld,et al.  A collaborative multiyear, multimodel assessment of seasonal influenza forecasting in the United States , 2019, Proceedings of the National Academy of Sciences.

[18]  Jiangzhuo Chen,et al.  DEFSI: Deep Learning Based Epidemic Forecasting with Synthetic Information , 2019, AAAI.

[19]  Jeremy Ginsberg,et al.  Detecting influenza epidemics using search engine query data , 2009, Nature.

[20]  Ronald Rosenfeld,et al.  Nonmechanistic forecasts of seasonal influenza with iterative one-week-ahead distributions , 2018, PLoS Comput. Biol..

[21]  Madhav V. Marathe,et al.  A framework for evaluating epidemic forecasts , 2017, BMC Infectious Diseases.

[22]  E. Dong,et al.  An interactive web-based dashboard to track COVID-19 in real time , 2020, The Lancet Infectious Diseases.

[23]  Naren Ramakrishnan,et al.  EpiDeep: Exploiting Embeddings for Epidemic Forecasting , 2019, KDD.

[24]  C. Murray.  Forecasting the impact of the first wave of the COVID-19 pandemic on hospital demand and deaths for the USA and European Economic Area countries , 2020, medRxiv.

[25]  P. Young,et al.  Time series analysis, forecasting and control , 1972, IEEE Transactions on Automatic Control.

[26]  Yu Wang,et al.  Curating a COVID-19 data repository and forecasting county-level death counts in the United States , 2020, Harvard Data Science Review.

[27]  Anil Vullikanti,et al.  Using data-driven agent-based models for forecasting emerging infectious diseases. , 2017, Epidemics.

[28]  Marc Lipsitch,et al.  Absolute humidity and pandemic versus epidemic influenza. , 2011, American journal of epidemiology.

[29]  Inderjit S. Dhillon,et al.  Clustering to forecast sparse time-series data , 2015, 2015 IEEE 31st International Conference on Data Engineering.

[30]  L. Wang,et al.  Epidemic Model Guided Machine Learning for COVID-19 Forecasts in the United States , 2020, medRxiv.

[31]  Daniela Perrotta,et al.  Forecasting Seasonal Influenza Fusing Digital Indicators and a Mechanistic Disease Model , 2017, WWW.