Using Hydrological Concepts and an Artificial Neural Network to Model the Rate for COVID-19 Infections versus Deaths
Hidraulica
; - (3):89-96, 2022.
Article
in English
| ProQuest Central | ID: covidwho-2045811
ABSTRACT
Models were run to reproduce COVID-19 infections versus deaths in Mexico City. The first model was made using rain runoff concept, emulating rain as number of infections reproducing runoff as number of deaths given as of March 2020. The second consisted of using an artificial neural network (ANN) proposed as an initial condition function to be implemented in the model with delay. These models were applied to fit accumulated confirmed case data, obtaining fit corroborated by coefficient of determination, R2. The R2 value produced by model was 0.0528 in case of infections comparison vs. official deaths reported by the Ministry of Health, 0.0571 for t case of infections vs. modelling using the HEC-HMS tool, and 0.0937 for case of contagion vs. modelling using ANN.
Engineering--Mechanical Engineering; Infections; Severe acute respiratory syndrome coronavirus 2; Runoff; Modelling; Artificial neural networks; Hydrology; COVID-19; Rain; Growth models; Precipitation; Pandemics; Neural networks; Coronaviruses; Fatalities; Soil conservation; United States--US; Mexico
Search on Google
Collection:
Databases of international organizations
Database:
ProQuest Central
Language:
English
Journal:
Hidraulica
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS