eVision: Epidemic Forecasting on COVID-19

Vaccination is the primary strategy to prevent COVID-19 illness and hospitalization. However, supplies are scarce and due to the regional mutations of the virus, new vaccines or booster shots will need to be administered potentially regularly. Hence, the prediction of the rate of growth of COVID-19 cases is paramount to ensuring the ample supply of vaccines as well as for local, state, and federal government measures to ensure the availability of hospital beds, supplies, and staff. eVision is an epidemic forecaster aimed at combining Machine Learning (ML) - in the form of a Long Short-Term Memory (LSTM) Recursive Neural Network (RNN) - and search engine statistics, in order to make accurate predictions about the weekly number of cases for highly communicable diseases. By providing eVision with the relative popularity of carefully selected keywords searched via Google along with the number of positive cases reported from the US Centers for Disease Control and Prevention (CDC) and/or the World Health Organization (WHO) the model can make highly accurate predictions about the trend of the outbreak by learning the relationship between the two trends. Thus, in order to predict the trend of the outbreak in a specific region, eVision is provided with a weekly count of the number of COVID-19 cases in a region along with statistics surrounding common symptom search phrases such as “loss of smell” and “loss of taste” that have been searched on Google in that region since the start of the pandemic. eVision has, for instance, been able to achieve an accuracy of %89 for predicting the trend of the COVID-19 outbreak in the United States

Agent-Based Simulation Framework for Epidemic Forecasting during Hajj Seasons in Saudi Arabia

The religious pilgrimage of Hajj is one of the largest annual gatherings in the world. Every year approximately three million pilgrims travel from all over the world to perform Hajj in Mecca in Saudi Arabia. The high population density of pilgrims in confined settings throughout the Hajj rituals can facilitate infectious disease transmission among the pilgrims and their contacts. Infected pilgrims may enter Mecca without being detected and potentially transmit the disease to other pilgrims. Upon returning home, infected international pilgrims may introduce the disease into their home countries, causing a further spread of the disease. Computational modeling and simulation of social mixing and disease transmission between pilgrims can enhance the prevention of potential epidemics. Computational epidemic models can help public health authorities predict the risk of disease outbreaks and implement necessary intervention measures before or during the Hajj season. In this study, we proposed a conceptual agent-based simulation framework that integrates agent-based modeling to simulate disease transmission during the Hajj season from the arrival of the international pilgrims to their departure. The epidemic forecasting system provides a simulation of the phases and rituals of Hajj following their actual sequence to capture and assess the impact of each stage in the Hajj on the disease dynamics. The proposed framework can also be used to evaluate the effectiveness of the different public health interventions that can be implemented during the Hajj, including size restriction and screening at entry points.

APPLICATION OF STOCHASTIC MODEL FOR LENGTHY EPIDEMIC FORECASTING

In this paper we consider a stochastic discrete-time epidemic model, with the infectivity depending on the age of infection and existing formula for the maximum likelihood estimation of the parameter responsible for the rate of the infection spread. In order to utilize the real number of infection cases statistics, a detection rate parameter is introduced. A program for automatic parameter estimation using past data with future epidemic simulation is developed. We present the comparison between the simulation of COVID-19 cases in Kyiv and real data using manual and automatic parameter estimation. We consider the possibility of the epidemic partition into several intervals with different parameters in order to simulate lengthy epidemics with significant changes in dynamics. We present the comparison between different numbers of partitions for long-term COVID-19 simulation in Kyiv (Ukraine) and Czech Republic, which have different dynamics of the epidemic development.

A Temporal Convolutional Recurrent Neural Network Model for Epidemic Forecasting Using Multi-Dimensional Data: A new method (Preprint)

BACKGROUND Infectious diseases such as COVID-19, influenza, Malaria, and Dengue have caused a significant threat throughout the world. For example, the expected yearly cost of pandemic influenza at roughly $500 billion, while COVID-19 has diminished the economic activity and could potentially lead to structural shifts in the global economy. One of the underlying major problems regarding these traditional surveillance epidemic methods is that they are not always effective and also the results produced by these methods usually have a delay of several weeks. OBJECTIVE The purpose of this study is to develop an epidemic forecasting model utilizing the deep learning technology that can be adapted to epidemic datasets and can predict the incidence or number of infectious diseases more accurately than traditional epidemic prediction methods. METHODS To predict the incidence of the epidemic, in this study, we collect real-world infectious disease data and transformed the dataset into time series. Our method uses the following information as inputs : (1) environmental and climatic information (2) epidemic–related internet search activity, (3) Google Trends, and (4) CDC ILI, Dengue Fever, Measles Incidence Data and related historical data. The proposed deep learning method utilizes a temporal convolutional technique that enables the exploitation of complex temporal patterns of epidemic activity across historical observations series. In the proposed deep learning model, we use the long and short-term memory units in a recurrent neural network to learn the temporal pattern of historical data. RESULTS We compare our model with three state-of-the-art deep learning models to evaluate the performance, accuracy, and relevance of the model predictions. We input the epidemic incidence data and observation data of the past 12 weeks to predict the number of incidence of the next one, two, and three weeks, respectively. We evaluate these models on the four real-world data sets we collected. The experiments demonstrate that our proposed model is better than several other models. CONCLUSIONS Previous studies often use autoregressive models or traditional machine learning methods to predict future epidemics. Compared to these, the performance evaluation of our method shows that our proposed method is superior to traditional non-machine methods and basic neural network models.

AutoSEIR: Accurate Forecasting from Real-time Epidemic Data Using Machine Learning

Since the SARS-CoV-2 virus outbreak has been recognized as a pandemic on March 11, 2020, several models have been proposed to forecast its evolution following the governments' interventions. In particular, the need for fine-grained predictions, based on real-time and fluctuating data, has highlighted the limitations of traditional SEIR models and parameter fitting, encouraging the study of new models for greater accuracy. In this paper we propose a novel approach to epidemiological parameter fitting and epidemic forecasting, based on an extended version of the SEIR compartmental model and on an auto-differentiation technique for partially observable ODEs (Ordinary Differential Equations). The results on publicly available data show that the proposed model is able to fit the daily cases curve with greater accuracy, obtaining also a lower forecast error. Furthermore, the forecast accuracy allows to predict the peak with an error margin of less than one week, up to 50 days before the peak happens.