Related Experiment Video
Updated: Jun 29, 2026
09:16
Supervised Machine Learning for Semi-Quantification of Extracellular DNA in Glomerulonephritis
Published on: June 18, 2020
Granular insights: A wastewater-based machine learning approach for localized COVID-19 hospitalization forecasting
Nusrat Tabassum1, Mohammad Mihrab Chowdhury1, Christopher S McMahan2
1Center for Public Health Modeling and Response, Clemson University, Clemson, SC, USA; Department of Public Health, Clemson University, Clemson, SC, USA.
Epidemics
|April 3, 2026
Summary
Wastewater-based epidemiology (WBE) accurately predicts COVID-19 hospitalizations up to 14 days in advance. This approach enhances public health surveillance and preparedness for infectious diseases.
Area of Science:
- Environmental science
- Epidemiology
- Public health
Background:
- Wastewater-based epidemiology (WBE) is crucial for monitoring community disease trends.
- Predicting hospitalizations aids healthcare resource management and preparedness.
Purpose of the Study:
- To evaluate SARS-CoV-2 RNA concentrations in wastewater for predicting COVID-19 hospitalizations in South Carolina.
- To assess the efficacy of WBE for fine-scale geographic predictions.
Main Methods:
- Analysis of SARS-CoV-2 RNA in wastewater from six treatment plants (April 2020 - February 2021).
- Utilized Poisson regression and random forest models to forecast 7, 14, and 21-day ahead hospitalizations.
- Validated model performance against statewide hospitalization claims data.
Main Results:
- Random forest models showed strongest accuracy for 14-day ahead predictions.
- Achieved median percentage agreement of 91.16% across wastewater treatment plants and 78.12% across ZIP codes.
- Demonstrated robust and timely prediction capabilities at fine geographic scales.
Conclusions:
- Wastewater-based epidemiology provides a reliable method for predicting infectious disease hospitalizations.
- The developed modeling framework can be adapted for surveillance of other infectious diseases.
- WBE enhances public health response and preparedness efforts.
