key: cord-0118381-4vgbzcx8 authors: Chae, Seo Yoon; Lee, Kyoung-Eun; Lee, Hyun Min; Jun, Nam; Le, Quang Ahn; Mafwele, Biseko Juma; Lee, Tae Ho; Kim, Doo Hwan; Lee, Jae Woo title: Estimation of Infection Rate and Prediction of Initial Infected Individuals of COVID-19 date: 2020-04-27 journal: nan DOI: nan sha: 2dbbd338a162539f945001a81e7ea3d1094ae436 doc_id: 118381 cord_uid: 4vgbzcx8 We consider the pandemic spreading of COVID-19 for some selected countries after the outbreak of the coronavirus in Wuhan City, China. We estimated the infection rate and the initial infected individuals of COVID-19 by using the officially reported data at the early stage of the epidemic for the susceptible (S), infectable (I), quarantined (Q), and the cofirmed recovered (Rk) population model, so called SIQRk model. In the reported data we know the quarantined cases and the recovered cases. We can not know the recovered cases from the asymptomatic cases. In the SIQRk model we can estimated the model parameters and the initial infecting cases (confirmed ans asymtomatic cases) from the data fits. We obtained the infection rate in the range between 0.233 and 0.462, the basic reproduction number Ro in the range between 1.8 and 3.5, and the initial number of infected individuals in the range betwee 10 and 8409 for some selected countries. By using fitting parameters we estimated the maximum time of the infection for Germany when the government are performing the quarantine policy. The disease is undergoing to the calm state about six months after first patients were identified. On China, the COVID-19 was spreading very quickly all over the world [2] and the COVID-19 is first pandemic disease in the 21 century. Some countries such as Republic of Korea, Taiwan, Singapore, Chinese Hong Kong are controlling the disease successfully up to now. However, other countries like USA, Italy, Spain, France, UK etc. are suffering the outbreak and the shortages of the medical materials and hospitals. After the outbreak the scientists all over the world are struggling to find out the vaccine and the treating drug. In the highly connected societies the information and data for the diseases are shared through the internet, social media, and mass media. We can obtain information from the website such as worldometer [3] or livecornamap in South The flooding articles and preprints are appearing on the many journals and preprint websites. Recently the preprint websites like arXiv.org [5] , bioRxiv [6], and medRxiv [7] are servicing the section for COVID-19 quick links. It is important to predict the spreading of the disease in the early stage of the outbreak. Many epidemic models are proposed based on the dynamic spreading models, agent-based models, Monte Carlo model, and databased spreading models [9] [10] [11] [12] [13] [14] . The evolution of the virus was described by the modified susceptible (S), infectious (I), recovered (R) population, so called SIR model. The prediction of COVID-19 evolution in Brazil has been suggested by the susceptible, infectious, quarantined, recovered (SIQR) model [9] . By the numerical analysis, he estimated the basis reproduction number In this model the parameter denotes the infection rate, the rate with which patients become non-infectious by the recovering without any symptoms. The parameter is the rate of detection of new infecting people, and is the recovering rate from the quarantined cases. In the SIQR k model, the infected populations are divided by the officially confirmed cases and the asymptomatic cases. We only know the official quarantined cases and the official recovered cases from the infection. We don't know the actual number of the infecting population due to the asymptomatic cases. The asymptomatic individuals are recovered without any severe suffering of the disease. The recovering population without symptoms are represented by ( ). We propose the parameters included in the dynamic equations and the initial number of infecting cases which is sum of the officially known cases and the unknown population of the asymptomatic cases. In the reported data of each countries, the active cases are transferring immediately to quarantined cases. Therefore, the active cases correspond to the quarantined case Q. Almost recovering cases is coming from the isolated cases. From the reported data for Q and R k we can fit (Q+R k ) as a function of time at the early stage of disease spreading. The in the range between 5% and 80% of the people testing positive for COVID-19 being asymptomatic. We set the fraction as = 1/3 [11] [12] [13] [14] [15] [16] [17] . The average incubation time is about 5 days [12, 13] and the duration of the milder cases of disease is about 5~6 days [14] . The average time of duration from infection to recovery or death of non-isolated cases are about 10 days, corresponding to a rate of 0.1/day [11] . Therefore, we obtain the rates as = (1 − ) × 0.1/day and = × 0.2/day. Finally, we obtain the parameters as = = 0.067/day. Using these parameters and the fitting parameter and , we obtain the parameter and (0) from the fitting parameters and the predetermined rates. We summarized the results obtained from the data of each countries. We estimated the infection rate and the initial number of the infected individuals (0). The symptoms of the COVID-19 are not appearing in many cases. In Fig. 1 we represent the nonlinear square fits of Q+R K as a function of time at the early stage of the disease spreading in Germany. The early data are well fitted by the exponential function. We give the fitting data for some selected countries in Table 2 . We observed that there were large number of initial infected people. The infection rate shows very high value in the range 0.233 ≤ ≤ 0.462 for the selected countries. We calculate the basic reproduction numbers of the estimated parameter for the countries. The of many countries is greater than 2. In particular, the basic reproduction number for the USA shows high value of = 3.45. This high value is inducing the large number of infecting people over the states in USA. We observed the high number of the initial infecting individuals (0) from the data fitting. In Table 1 we summarized the first official confirmed day of the COVID-19 patient. Because of the incubating period and asymptomatic cases for young health people, we Fig. 2 . When we predict the evolution of the disease by some model, we required to use the confirmed data set such as the active cases, recovered cases, and death cases. 10 We consider an epidemic spreading model, SIQR K model. In this model we include the dynamic equation for the quarantined individuals. We estimated the parameters of the dynamic evolution equation from the sum of the quarantined cases and the recovered cases. We obtained the parameters by the nonlinear least square fits by using the reported data set. It is very important that we consider the asymptomatic individuals when we predict the dynamic evolution of the disease by some model. The observed high value of the basic reproduction number indicated the huge pandemic of the disease all over the world. We predict that the maximum time of the infection is around 50 days or two months. The disease should be lasting about six months when we have quarantined the infecting individuals. Let's solve the SIQR K model. In the early phase of the disease spreading we expect that the susceptible population is similar to the total population / ≈ 1. Therefore, we can write the infection dynamic equation such as [10, 11] The reproduction number is given by In the COVID-19, the reproduction number is bigger than one. The disease can spread easily by the contact process between individuals. The double time is given by = We calculate the recovering rate obtained by the data set. The recovering rate is given by = ( − −1 )/ −1 . The value of the recovering rate depends on the time at the early stage and converge to a constant value. We obtained the recovering rate as = 0.036/day. Novel Coronavirus (2019-nCoV) situation report-1, World Health Organization WHO Director-General's opening remarks at the media briefing on COVID-19 -11 Epidemic analysis of COVID-19 in China by dynamical modeling A simulation of a COVID-19 epidemic based on a deterministic SEIR model A simulation of a COVID-19 epidemic based on a deterministic SEIR model Quantifying undetected COVID-19 cases and effects of containment measures in Italy Early Transmission dynamics in Wuhan, China, of Novel Coronavirus-Infected Pneumonia The Incubation Period of Coronavirus Disease 2019 (COVID-19) From Publicly eported Con_rmed Cases: Estimation and Application Presumed asymptomatic carrier transmission of COVID-19