Convolutional neural networks and temporal CNNs for COVID-19 forecasting in France

Abstract

This paper focus on multiple CNN-based (Convolutional Neural Network) models for COVID-19 forecast developed by our research team during the first French lockdown. In an effort to understand and predict both the epidemic evolution and the impacts of this disease, we conceived models for multiple indicators: daily or cumulative confirmed cases, hospitalizations, hospitalizations with artificial ventilation, recoveries, and deaths. In spite of the limited data available when the lockdown was declared, we achieved good short-term performances at the national level with a classical CNN for hospitalizations, leading to its integration into a hospitalizations surveillance tool after the lockdown ended. Also, A Temporal Convolutional Network with quantile regression successfully predicted multiple COVID-19 indicators at the national level by using data available at different scales (worldwide, national, regional). The accuracy of the regional predictions was improved by using a hierarchical pre-training scheme, and an efficient parallel implementation allows for quick training of multiple regional models. The resulting set of models represent a powerful tool for short-term COVID-19 forecasting at different geographical scales, complementing the toolboxes used by health organizations in France.

Nation-wide human mobility prediction based on graph neural networks

“…In this context, one of the major findings in this discipline is that human mobility is highly predictable [3].Therefore, many techniques and solutions have emerged for human mobility forecasting [4]. Moreover, the anticipation of human flow has important applications in diverse domains like healthcare [5,6], urban services [7] and transportation management [8].…”

“…Fernando Terroso-Saenz fterroso@ucam.edu 1 UCAM, Campus de los Jerónimos, Guadalupe 30107 Murcia, España These solutions often rely on human movement trajectories on different spatial and temporal scales from a large variety of sources like GPS traces [9], Call Detail Records (CDRs) [10] or Online Social Media (OSM) posts [11]. To analyze these sources, different techniques based on statistical methods [12], machine learning algorithms [13] and, more recently, deep learning [6] have been used.…”

A Comprehensive Review of Artificial Intelligence in Prevention and Treatment of COVID-19 Pandemic

“…Oliveira et al ( de Oliveira et al, 2021 ) proposed an Artificial Neural Network model, in which an ANN model was applied to predict the number of confirmed COVID-19 cases and deaths, as well as the time series for the next 7 days in Brazil, Portugal and the United States. Mohimont et al ( Mohimont and Chemchem, 2020 ) mainly studied a number of models based on CNN, and also proposed a layered transfer learning scheme. Finally, good national and regional accuracy is obtained, and the performance of ordinary CNN is improved.…”

Simulating Immunization Campaigns and Vaccine Protection Against COVID-19 Pandemic in Brazil

“…An effective way to assess and forecast the number of cases and deaths by COVID-19 is by applying deep learning, such as Artificial Neural Networks (ANN). Most recently, popular deep learning architectures like Recursive Neural Networks [28]- [32] and Convolutional Neural Networks [33], [34] have been successfully used for forecasting COVID-19 timeseries without the inclusion of compartmental models of infections dynamics [35]. However, concerning the vaccination data, the fresh literature on purely ANN-based methods is very scarce [36].…”