Physics-informed machine learning for the COVID-19 pandemic: Adherence to social distancing and short-term predictions for eight countries

G. Barmparis, G. Tsironis

Like Comment

Received date: 20th August 2020

The analysis of COVID-19 infection data through the eye of Physics-inspired Artificial Intelligence leads to a clearer understand- ing of the infection dynamics and assists in predicting future evolution. The spreading of the pandemic during the first half of 2020 was curtailed to a larger or lesser extent through social distancing measures imposed by most countries. In the context of the standard Susceptible-Infected-Recovered model, changes in social distancing enter through time-dependent infection rates. The use of machine learning enables to extract from the infection data the degree of social distancing and, through it, assess the effectiveness of the imposed measures. We perform this quantitative analysis for eight countries with infection data from the first viral wave. We find as two extremes Greece and USA where the measures were successful and unsuccessful, respectively, in limiting spreading during this phase. Subsequently, we employ this physics-based neural network approach to the second wave of the infection and train the network with the new data. Extraction of the time-dependent infection rate is performed through the trained neural network and, subsequently, it is used to make short-term predictions with a week-long or even longer horizon. We apply this algorithmic approach to all eight countries with good short-term results. The data for Greece is analyzed in more detail for the period of approximately four months, i.e., from August to December 2020. We find that the model captures the essential spreading dynamics and has reasonably good predictive power both in the short-term but also for more intermediate horizon. In particular, it gives, useful projections for the spreading based on specific social distancing measures that are extracted directly from the data.

Read in full at arXiv.

This is an abstract of a preprint hosted on a preprint server, which is currently undergoing peer review at Scientific Reports. The findings have yet to be thoroughly evaluated, nor has a decision on ultimate publication been made. Therefore, the results reported should not be considered conclusive, and these findings should not be used to inform clinical practice, or public health policy, or be promoted as verified information.

Scientific Reports

-, Nature Portfolio, Springer Nature