Early Prediction of the 2019 Novel Coronavirus Outbreak in the Mainland China Based on Simple Mathematical Model

Abstract

The 2019 novel coronavirus (2019-nCoV) outbreak has been treated as a Public Health Emergency of International Concern by the World Health Organization. This work made an early prediction of the 2019-nCoV outbreak in China based on a simple mathematical model and limited epidemiological data. Combing characteristics of the historical epidemic, we found part of the released data is unreasonable. Through ruling out the unreasonable data, the model predictions exhibit that the number of the cumulative 2019-nCoV cases may reach 76,000 to 230,000, with a peak of the unrecovered infectives (22,000-74,000) occurring in late February to early March. After that, the infected cases will rapidly monotonically decrease until early May to late June, when the 2019-nCoV outbreak will fade out. Strong anti-epidemic measures may reduce the cumulative infected cases by 40%-49%. The improvement of medical care can also lead to about one-half transmission decrease and effectively shorten the duration of the 2019-nCoV.INDEX TERMS Epidemic transmission, infection rate, mathematical model, novel coronavirus, prediction, removal rate.

SimCOVID: Open-Source Simulation Programs for the COVID-19 Outbreak

“…All the numerical data for the test cases were taken from [16]. The collected data (officially reported cases) displays some inconvenience observations in the reported cases of infection (the evidence is the jump in the curve on Feb. 13 th ) which is reportedly [6] caused by a sudden change in the way the data was collected.…”

“…Note that the cumulative infectious is the sum of the daily infectious values which can be modeled in Simulink using an integrator. As mentioned in [6], the sudden jump in the curve is due to the way the data was collected and the diagnosis techniques which is not interpreted by a natural variation of a pandemic. Therefore, the program tries to change the parameters so that the simulated cumulative cases are close to the actual recorded ones.…”

“…An essential part of minimizing the spread of the virus is to monitor, track, and estimate the final size of the pandemic. This is extremely useful for decision making against the public health crises [6]. One way to predict the dynamic spread of the epidemic is by the use of computer simulation following the mathematical model of an epidemic.…”

“…One way to predict the dynamic spread of the epidemic is by the use of computer simulation following the mathematical model of an epidemic. In the literature, several analytical approaches have been proposed to model the pandemic including Susceptible-Infected-Removed (SIR) model [6][7], Susceptible-Exposed-Infected-Removed (SEIR) model [1], Susceptible-Infected-Recovered-Dead (SIRD) model [8,9], and fractional-derivative SEIR [10], and SEIRD [11]. While some recent studies are addressing this epidemic using the aforementioned models [6][7][8][9][10][11], there is an increasing need to develop an open-source computer program to perform a time-domain simulation of the dynamic spread of the virus.…”

COVID-19 Outbreak Prediction with Machine Learning

“…Often to acquire information on class R, several novel models included data from social media or call data records (CDR), which showed promising results [18][19][20][21][22][23][24][25]. However, observation of the behavior of COVID-19 in several countries demonstrates a high degree of uncertainty and complexity [26]. Thus, for epidemiological models to be able to deliver reliable results, they must be adapted to the local situation with an insight into susceptibility to infection [27].…”