Researchers have developed a machine learning framework to help water managers identify viral and bacterial contamination risks before conventional laboratory testing is complete, combining environmental data with quantitative microbial risk assessment (QMRA). The study, published in *Biocontaminant*, addresses the challenge that traditional fecal contamination indicators—fecal coliforms, *Escherichia coli*, and *Enterococcus faecalis*—do not always accurately reflect the presence of viruses, which can behave differently in aquatic environments. The researchers collected 95 surface-water samples from two drinking-water sources in a major city in eastern China between May 2024 and December 2025. Each sample was analyzed for the three bacterial indicators and six pathogens: *Pseudomonas aeruginosa*, *Salmonella* species, *Shigella* species, adenovirus, norovirus, and enterovirus. The three fecal indicator bacteria were significantly correlated with one another, but their relationships with viral pathogens were generally weak or inconsistent, highlighting a limitation in relying exclusively on bacterial indicators. The team compared six machine learning methods—multiple linear regression, least-squares boosting, decision trees, support vector machines, random forests, and multilayer perceptrons—to determine how effectively physicochemical measurements could predict pathogen concentrations.