Computational Forecasting Methodology for Acute Respiratory Infectious Disease Dynamics.

Daniel Alejandro Gónzalez-Bandala,Daniel E Noyola,Christian A García-Sepúlveda,Juan Carlos Cuevas-Tello,Andreu Comas-García

doi:10.3390/ijerph17124540

Abstract

The study of infectious disease behavior has been a scientific concern for many years as early identification of outbreaks provides great advantages including timely implementation of public health measures to limit the spread of an epidemic. We propose a methodology that merges the predictions of (i) a computational model with machine learning, (ii) a projection model, and (iii) a proposed smoothed endemic channel calculation. The predictions are made on weekly acute respiratory infection (ARI) data obtained from epidemiological reports in Mexico, along with the usage of key terms in the Google search engine. The results obtained with this methodology were compared with state-of-the-art techniques resulting in reduced root mean squared percentage error (RMPSE) and maximum absolute percent error (MAPE) metrics, achieving a MAPE of 21.7%. This methodology could be extended to detect and raise alerts on possible outbreaks on ARI as well as for other seasonal infectious diseases.

Highlights

Acute respiratory infections (ARI) are one of the main causes of morbidity and mortality in the world, in children under 5 years old and adults over 65 years old
This paper presents a methodology capable of making accurate predictions of ARI activity with data obtained from epidemiological reports along with search terms usage derived from the Google search engine
The recorded results are focused on the last stage of the methodology, the computational model, i.e., forecasting model (FFNN), projection model (SoS), and the merge prediction

Summary

Introduction

Acute respiratory infections (ARI) are one of the main causes of morbidity and mortality in the world, in children under 5 years old and adults over 65 years old. The most frequent pathogens that cause ARI are respiratory syncytial virus (RSV), human metapneumovirus, rhinovirus/enterovirus, influenza viruses, parainfluenza 1–4, adenovirus, coronavirus, Streptococcus pneumoniae, and Mycoplasma pneumoniae [8,9]. ARI exhibit a seasonal pattern where RSV and influenza viruses are the major contributing pathogens. Changes in circulating viral strains of these viruses may result in yearly winter ARI epidemics. The introduction of novel influenza strains or other viruses into the human population can lead to the emergence of pandemics

Methods

Results

Discussion

Conclusion