Quantifying the effect of quarantine control in Covid-19 infectious spread using machine learning

Abstract

Since the first recording of what we now call Covid-19 infection in Wuhan, Hubei province, China on Dec 31, 2019, the disease has spread worldwide and met with a wide variety of social distancing and quarantine policies. The effectiveness of these responses is notoriously difficult to quantify as individuals travel, violate policies deliberately or inadvertently, and infect others without themselves being detected. Moreover, the publicly available data on infection rates are themselves unreliable due to limited testing and even possibly under-reporting. In this paper, we attempt to interpret and extrapolate from publicly available data using a mixed first-principles epidemiological equations and data-driven neural network model. Leveraging our neural network augmented model, we focus our analysis on four locales: Wuhan, Italy, South Korea and the United States of America, and compare the role played by the quarantine and isolation measures in each of these countries in controlling the effective reproduction number Rt of the virus. Our results unequivocally indicate that the countries in which rapid government interventions and strict public health measures for quarantine and isolation were implemented were successful in halting the spread of infection and prevent it from exploding exponentially. In the case of Wuhan especially, where the available data were earliest available, we have been able to test the predicting ability of our model by training it from data in the January 24 till March 3 window, and then matching the predictions up to April 1. Even for Italy and South Korea, we have a buffer window of one week (25 March - 1 April) to validate the predictions of our model. In the case of the US, our model captures well the current infected curve growth and predicts a halting of infection spread by 20 April 2020. We further demonstrate that relaxing or reversing quarantine measures right now will lead to an exponential explosion in the infected case count, thus nullifying the role played by all measures implemented in the US since mid March 2020.

Spread of Covid-19 in the United States is controlled

“…The single layer AE attempts to minimize the error between the input vector and the reconstruction vector. We develop stacked autoencoders with 4 layers that consist of two single-layer AEs stacked layer by layer (2). The dimensions of the input layer, the first hidden layer and the second hidden layer are 8, 32 and 4, respectively ( Figure S1(a)).…”

ARIMA Models for Predicting the End of COVID-19 Pandemic and the Risk of a Second Rebound

“…Raj Dandekar and George Barbastathis [14] proposed a method to capture the current infected curve growth and predicts a halting of infection spread by 20 April 2020. This method has shown that reversing quarantine measures right at this time can lead to an exponential explosion in the infected case count, thus annulling the part played by all measures implemented in the US since mid-March 2020.…”

Simplified model of Covid-19 epidemic prognosis under quarantine and estimation of quarantine effectiveness

“…Consequently, on the basis of eq. ( 3 ), the parameter of SIR model can be estimated as the number of new registered cases of infection to number of active cases ratio: ( 4 ) where is the rate of new infected cases ( 5 ) which can be estimated by counting new cases of infection, and usually is measured by number of registered new cases per time period ( 6 ) The number of daily new infected cases is defined as ( 7 ) So, expected number of new cases in next day could be predicted by the today number of active infected individuals multiplied by the infectious rate ( 8 ) In general, the infectious rate is time dependent and usually can be described by a complex function with additional parameters that must be daily calibrated according to last epidemiological data. Behaviour of the infectious rate during quarantine may differ in different countries, which reduces possibilities to build correct model of the infectious rate on the basis of epidemiological data from other countries.…”

“…Valuable information that could be obtained from modelling is forecast of the expected time and number of most active infected cases and the effectiveness of applied infection control measures. It is current global trend, that the experience and available data from already affected countries are used to model the pandemic dynamics in other countries before the epidemic has reached the peak or to estimate effectiveness of various scenario of the next wave management [4].…”