Application of machine learning for multi-community COVID-19 outbreak predictions with wastewater surveillance.

The potential of wastewater-based epidemiology (WBE) as a surveillance and early warning tool for the COVID-19 outbreak has been demonstrated. For areas with limited testing capacity, wastewater surveillance can provide information on the disease dynamic at a community level. A predictive model is a key to generating quantitative estimates of the infected population. Modeling longitudinal wastewater data can be challenging as biomarkers in wastewater are susceptible to variations caused by multiple factors associated with the wastewater matrix and the sewersheds characteristics. As WBE is an emerging trend, the model should be able to address the uncertainties of wastewater from different sewersheds. We proposed exploiting machine learning and deep learning techniques, which are supported by the growing WBE data. In this article, we reviewed the existing predictive models, among which the emerging machine learning/deep learning models showed great potential. However, most models are built for individual sewersheds with few features extracted from the wastewater. To fulfill the research gap, we compared different time-series and non-time-series models for their short-term predictive performance of COVID-19 cases in 9 diverse sewersheds. The time-series models, long short-term memory (LSTM) and Prophet, outcompeted the non-time-series models. Besides viral (SARS-CoV-2) loads and location identity, domain-specific features like biochemical parameters of wastewater, geographical parameters of the sewersheds, and some socioeconomic parameters of the communities can contribute to the models. With proper feature engineering and hyperparameter tuning, we believe machine learning models like LSTM can be a feasible solution for the COVID-19 trend prediction via WBE. Overall, this is a proof-of-concept study on the application of machine learning in COVID-19 WBE. Future studies are needed to deploy and maintain the model in more real-world applications.

Socioeconomic status correlations with confirmed COVID-19 cases and SARS-CoV-2 wastewater concentrations in small-medium sized communities.

Over two years into the COVID-19 pandemic, it is apparent that some populations across the world are more susceptible than others to SARS-CoV-2 infection and spread. Understanding how populations with varying demographic patterns are impacted by COVID-19 may highlight which factors are most important in targeting to combat global suffering. The first objective of this study was to investigate the association of various socioeconomic status (SES) parameters and confirmed COVID-19 cases in the state of Ohio, USA. This study examines the largest and capital city of Ohio (Columbus) and various small-medium-sized communities. The second objective was to determine the relationship between SES parameters and community-level SARS-CoV-2 concentrations using municipal wastewater samples from each city's respective wastewater treatment plants from August 2020 to January 2021. SES parameters include population size, median income, poverty, race/ethnicity, education, health care access, types of COVID-19 testing sites, and social vulnerability index. Statistical analysis results show that confirmed (normalized and/or non-normalized) COVID-19 cases were negatively associated with White percentage and registered hospitals, and positively associated with registered physicians and various COVID-19 testing sites. Wastewater viral concentrations were negatively associated with poverty, and positively associated with median income, community health centers, and onsite rapid testing locations. Additional analyses conclude that population is a significant factor in determining COVID-19 cases and SARS-CoV-2 wastewater concentrations. Results indicate that community healthcare parameters relate to a negative health outcome (COVID-19) and that demographic parameters can be associated with community-level SARS-CoV-2 wastewater concentrations. As the first study that examines the association between socioeconomic parameters and SARS-CoV-2 wastewater concentrations as well as confirmed COVID-19 cases, it is apparent that social determinants have an impact in determining the health burden of small-medium sized Ohioan cities. This study design and innovative approach are scalable and applicable for endemic and pandemic surveillance across the world.

Wastewater surveillance of SARS-CoV-2 in dormitories as a part of comprehensive university campus COVID-19 monitoring.

Wastewater-based epidemiology is an effective tool for monitoring infectious disease spread or illicit drug use within communities. At the Ohio State University, we conducted a SARS-CoV-2 wastewater surveillance program in the 2020-2021 academic year and compared results with the university-required weekly COVID-19 saliva testing to monitor COVID-19 infection prevalence in the on-campus residential communities. The objectives of the study were to rapidly track trends in the wastewater SARS-CoV-2 gene concentrations, analyze the relationship between case numbers and wastewater signals when adjusted using human fecal viral indicator concentrations (PMMoV, crAssphage) in wastewater, and investigate the relationship of the SARS-CoV-2 gene concentrations with wastewater parameters. SARS-CoV-2 nucleocapsid and envelope (N1, N2, and E) gene concentrations, determined with reverse transcription droplet digital PCR, were used to track SARS-CoV-2 viral loads in dormitory wastewater once a week at 6 sampling sites across the campus during the fall semester in 2020. During the following spring semester, research was focused on SARS-CoV2 N2 gene concentrations at 5 sites sampled twice a week. Spearman correlations both with and without adjusting using human fecal viral indicators showed a significant correlation (p < 0.05) between human COVID-19 positive case counts and wastewater SARS-CoV-2 gene concentrations. Spearman correlations showed significant relationships between N1 gene concentrations and both TSS and turbidity, and between E gene concentrations and both pH and turbidity. These results suggest that wastewater signal increases with the census of infected individuals, in which the majority are asymptomatic, with a statistically significant (p-value <0.05) temporal correlation. The study design can be utilized as a platform for rapid trend tracking of SARS-CoV-2 variants and other diseases circulating in various communities.

Wastewater SARS-CoV-2 monitoring as a community-level COVID-19 trend tracker and variants in Ohio, United States.

The global pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has resulted in more than 129 million confirm cases. Many health authorities around the world have implemented wastewater-based epidemiology as a rapid and complementary tool for the COVID-19 surveillance system and more recently for variants of concern emergence tracking. In this study, three SARS-CoV-2 target genes (N1 and N2 gene regions, and E gene) were quantified from wastewater influent samples (n = 250) obtained from the capital city and 7 other cities in various size in central Ohio from July 2020 to January 2021. To determine human-specific fecal strength in wastewater samples more accurately, two human fecal viruses (PMMoV and crAssphage) were quantified to normalize the SARS-CoV-2 gene concentrations in wastewater. To estimate the trend of new case numbers from SARS-CoV-2 gene levels, different statistical models were built and evaluated. From the longitudinal data, SARS-CoV-2 gene concentrations in wastewater strongly correlated with daily new confirmed COVID-19 cases (average Spearman's r = 0.70, p < 0.05), with the N2 gene region being the best predictor of the trend of confirmed cases. Moreover, average daily case numbers can help reduce the noise and variation from the clinical data. Among the models tested, the quadratic polynomial model performed best in correlating and predicting COVID-19 cases from the wastewater surveillance data, which can be used to track the effectiveness of vaccination in the later stage of the pandemic. Interestingly, neither of the normalization methods using PMMoV or crAssphage significantly enhanced the correlation with new case numbers, nor improved the estimation models. Viral sequencing showed that shifts in strain-defining variants of SARS-CoV-2 in wastewater samples matched those in clinical isolates from the same time periods. The findings from this study support that wastewater surveillance is effective in COVID-19 trend tracking and provide sentinel warning of variant emergence and transmission within various types of communities.