What is it about?
Reliable monitoring of COVID-19 remains a public health priority. However, declining rates of routine testing of infection have resulted in many infection cases going unreported. By measuring community-level COVID-19 spread through target viruses in wastewater, we detect more reliable signals of community infection levels in comparison to traditional at-clinic tests. We built a machine-learning model to predict expected case counts given observed virus levels in wastewater samples collected from a community under monitoring during the period when infection case data were mandatorily reported. Such a model was then applied to the period when testing requirements were lifted, which consistently predicted more cases than reported in the CDC databases, suggesting an issue of significant systematic underreporting. Our findings demonstrate that wastewater-based community surveillance combined with machine-learning tools can provide more accurate information of disease prevalence when at-clinic testing is voluntary or otherwise incomplete. This novel wastewater monitoring system empowered by reliable prediction machinery can aid public health officials not only to detect outbreaks earlier but also to more reliably capture real-time infection conditions, so that it assists the government to make more informed decisions regarding disease control policy, especially when traditional at-clinic testing data prove limited or unreliable.
Featured Image
Photo by George Pagan III on Unsplash
Why is it important?
Wastewater monitoring systems empowered by reliable prediction machinery can aid public health officials not only to detect outbreaks earlier but also to more reliably capture real-time infection conditions, so that it assists the government to make more informed decisions regarding disease control policy, especially when traditional at-clinic testing data prove limited or unreliable.
Perspectives
This is a pioneer work that shows the way of borrowing machine learning techniques to empower wastewater surveillance systems in detecting disease outbreaks and removing errors of monitoring data.
Professor Peter X.K. Song
University of Michigan
Read the Original
This page is a summary of: Community‐level wastewater surveillance with machine learning methods to assess underreporting of COVID‐19 case counts, mLife, December 2025, Tsinghua University Press,
DOI: 10.1002/mlf2.70055.
You can read the full text:
Contributors
The following have contributed to this page
