Wastewater Surveillance for Early Detection: A Comparative Analysis of Transmission Dynamic Models.

Bong-Jin Choi Co-Author
North Dakota State University
 
Bong-Jin Choi Presenting Author
North Dakota State University
 
Sunday, Aug 4: 4:35 PM - 4:50 PM
3845 
Contributed Papers 
Oregon Convention Center 
There has been extensive research conducted on the transmission dynamics of COVID-19 disease. The SARS-CoV-2 virus primarily spreads through the respiratory tract. It is very common for the virus to spread rapidly during the incubation period. Further, asymptomatic carriers contribute to this rapid transmission. As an early detection method, wastewater surveillance can be used to detect viruses before they spread far and wide. Our study focused on collecting wastewater samples from treatment plants across various cities in North Dakota. Utilizing viral RNA copies, we compared the model predictions of K-Nearest Neighbor (KNN) regression, Quantile Regression (QR), and Long-Short-Term-Memory (LSTM) network models. To gauge its efficacy, we compared our models' predictions with those of the fundamental Susceptible-Infected-Recovered (SIR) model.

Keywords

SARS-CoV-2

K-Nearest Neighbor

Quantile Regression

Long-Short-Term-Memory

SIR model 

Main Sponsor

Section on Statistics in Epidemiology