An exploration of challenges associated with machine learning for time series forecasting of COVID-19 community spread using wastewater-based epidemiological data.

Wastewater-based epidemiology (WBE) has gained increasing attention as a complementary tool to conventional surveillance methods with potential for significant resource and labour savings when used for public health monitoring. Using WBE datasets to train machine learning algorithms and develop predictive models may also facilitate early warnings for the spread of outbreaks. The challenges associated with using machine learning for the analysis of WBE datasets and timeseries forecasting of COVID-19 were explored by running Random Forest (RF) algorithms on WBE datasets across 108 sites in five regions: Scotland, Catalonia, Ohio, the Netherlands, and Switzerland. This method uses measurements of SARS-CoV-2 RNA fragment concentration in samples taken at the inlets of wastewater treatment plants, providing insight into the prevalence of infection in upstream wastewater catchment populations. RF's forecasting performance at each site was quantitatively evaluated by determining mean absolute percentage error (MAPE) values, which was used to highlight challenges affecting future implementations of RF for WBE forecasting efforts. Performance was generally poor using WBE datasets from Catalonia, Scotland, and Ohio with 'reasonable' or better forecasts constituting 0 %, 5 %, and 0 % of these regions' forecasts, respectively. RF's performance was much stronger with WBE data from the Netherlands and Switzerland, which provided 55 % and 45 % 'reasonable' or better forecasts respectively. Sampling frequency and training set size were identified as key factors contributing to accuracy, while inclusion of too many unnecessary variables (or e.g., flow data) was identified as a contributing factor to poor performance. The contribution of catchment population on forecast accuracy was more ambiguous. This study determined that the factors governing RF's forecast performance are complicated and interrelated, which presents challenges for further work in this space. A sufficiently accurate further iteration of the tool discussed within this study would provide significant but varying value for public health departments for monitoring future, or ongoing outbreaks, assisting the implementation of on-time health response measures.

Show us the data: global COVID-19 wastewater monitoring efforts, equity, and gaps.

A year since the declaration of the global coronavirus disease 2019 (COVID-19) pandemic, there were over 110 million cases and 2.5 million deaths. Learning from methods to track community spread of other viruses such as poliovirus, environmental virologists and those in the wastewater-based epidemiology (WBE) field quickly adapted their existing methods to detect SARS-CoV-2 RNA in wastewater. Unlike COVID-19 case and mortality data, there was not a global dashboard to track wastewater monitoring of SARS-CoV-2 RNA worldwide. This study provides a 1-year review of the "COVIDPoops19" global dashboard of universities, sites, and countries monitoring SARS-CoV-2 RNA in wastewater. Methods to assemble the dashboard combined standard literature review, Google Form submissions, and daily, social media keyword searches. Over 200 universities, 1400 sites, and 55 countries with 59 dashboards monitored wastewater for SARS-CoV-2 RNA. However, monitoring was primarily in high-income countries (65%) with less access to this valuable tool in low- and middle-income countries (35%). Data were not widely shared publicly or accessible to researchers to further inform public health actions, perform meta-analysis, better coordinate, and determine equitable distribution of monitoring sites. For WBE to be used to its full potential during COVID-19 and beyond, show us the data.

Reference and point-of-care testing for G6PD deficiency: Blood disorder interference, contrived specimens, and fingerstick equivalence and precision.

Certain clinical indications and treatments such as the use of rasburicase in cancer therapy and 8-aminoquinolines for Plasmodium vivax malaria treatment would benefit from a point-of-care test for glucose-6-phosphate dehydrogenase (G6PD) deficiency. Three studies were conducted to evaluate the performance of one such test: the STANDARD™ G6PD Test (SD BIOSENSOR, South Korea). First, biological interference on the test performance was evaluated in specimens with common blood disorders, including high white blood cell (WBC) counts. Second, the test precision on fingerstick specimens was evaluated against five individuals of each, deficient, intermediate, and normal G6PD activity status. Third, clinical performance of the test was evaluated at three point-of-care settings in the United States. The test performed equivalently to the reference assay in specimens with common blood disorders. High WBC count blood samples resulted in overestimation of G6PD activity in both the reference assay and the STANDARD G6PD Test. The STANDARD G6PD Test showed good precision on multiple fingerstick specimens from the same individual. The same G6PD threshold values (U/g Hb) were applied for a semiquantitative interpretation for fingerstick- and venous-derived results. The sensitivity/specificity values (95% confidence intervals) for the test for G6PD deficiency were 100 (92.3-100.0)/97 (95.2-98.2) and 100 (95.7-100.0)/97.4 (95.7-98.5) for venous and capillary specimens, respectively. The same values for females with intermediate (> 30% to ≤ 70%) G6PD activity were 94.1 (71.3-99.9)/88.2 (83.9-91.7) and 82.4 (56.6-96.2)/87.6(83.3-91.2) for venous and capillary specimens, respectively. The STANDARD G6PD Test enables point-of-care testing for G6PD deficiency.