Epidemic Dynamics via Wavelet Theory and Machine Learning with Applications to Covid-19

Abstract

We introduce the concept of epidemic-fitted wavelets which comprise, in particular, as special cases the number I(t) of infectious individuals at time t in classical SIR models and their derivatives. We present a novel method for modelling epidemic dynamics by a model selection method using wavelet theory and, for its applications, machine learning-based curve fitting techniques. Our universal models are functions that are finite linear combinations of epidemic-fitted wavelets. We apply our method by modelling and forecasting, based on the Johns Hopkins University dataset, the spread of the current Covid-19 (SARS-CoV-2) epidemic in France, Germany, Italy and the Czech Republic, as well as in the US federal states New York and Florida.

Breast Cancer Surgery 10-Year Survival Prediction by Machine Learning: A Large Prospective Cohort Study

“…Compared to the other models, the superior forecasting performance of the DNN model and other advantages resulting from its unique characteristics are well established in the literature and are well supported by comprehensive statistical analyses and comparisons in previous works [ 19 , 20 , 21 ]. One advantage of the DNN model is its ability to process incomplete or noisy inputs more appropriately and more accurately compared to other models when no missing data occurs in the dataset.…”

A comparative study for predictive monitoring of COVID-19 pandemic

“…Hybrid models designed using compartmental models and deep learning frameworks have also been proposed recently [18] . Compartmental models are based on the assumption that the chance of an infected person to infect another susceptible person is constant during the epidemic duration and also, every infected person has a constant chance to recover at any given time, which might not be true [19] .…”

“…Simple curve-fitting approaches typically support parameter estimation of a single wave characterized by a single peak throughout the epidemic duration. However, fitting the data with only one wave may be incorrect since, in general, there are several recurring waves that emerge and die throughout the epidemic duration [22] . To overcome this drawback, some recent works have decomposed the available data into multiple overlapping waves, where every single wave is a generalized growth model such as the logistic or the Gaussian growth models [7] , [8] , [14] .…”

On fluctuating characteristics of global COVID-19 cases and identification of inflection points

“…G. Rohith et al modeled the dynamics of COVID-19 using the susceptible-exposed-fectious-removed model with a nonlinear incidence (Rohith and Devika, 2020). Kinetic fitting research on the transmission of the COVID-19 epidemic (Tat Dat et al , 2020; Agosto and Giudici, 2020) has been a major research aspect since the outbreak of the epidemic.…”