Artificial intelligence can accurately predict hospital admission numbers due to COVID-19 up to four weeks in advance, new research showed.
The research involved developing an AI-based system using COVID wastewater sampling from 159 countries in the US, covering nearly 100 million Americans, along with US hospital admission records to build the prediction model.
"Current prediction methods are based on COVID-19 laboratory testing, or self-testing and reporting, however this does not pick up asymptomatic cases, and many countries are moving away from rigorous testing requirements," said Dr. Xuan Li from the University of Technology Sydney's (UTS) Faculty of Engineering and IT.
Li together with colleague Professor Qilin Wang led the research alongside researchers from UNSW Sydney, Delft University of Technology and Morgan State University.
The study was supported by the Australian Research Council and the Australian Academy of Science.
"COVID-19 still poses a heavy burden on healthcare systems around the world. The number of Australians in hospital with COVID-19 peaked at around 5,500. Rapid increases in patient numbers can stress frontline healthcare capacity and increase fatality rates," said Li.
Dr. Xuan Li, UTS
Li was recently awarded a two-year grant from the Australian Academy of Science WH Gladstones Population and Environment Fund to develop an Australian-based wastewater prediction model.
Cost-effective early warning system
Professor Wang said the research reveals wastewater surveillance combined with AI-based modelling can be a cost-effective early warning system, allowing public health officials to better prepare for and manage pandemic waves, and efficiently allocate limited healthcare resources.
Wang noted that while wastewater monitoring is already conducted in many countries, it is limited to showing whether COVID-19 is present in a region, as well as a rough estimation of whether the burden is increasing or decreasing.
"We used AI to pick up patterns and changes in the data and learn from this to increase the accuracy of predictions.”
Professor Qilin Wang UTS
"Variables that can influence hospital admissions include changing behaviour due to public policies, vaccination rates, holidays and weather. The established model can help accurately predict the hospitalisation needs due to COVID-19 in the region," said Wang.
The way forward
Li hopes to extend her research to include other infectious diseases that can be detected through wastewater-based epidemiology, including food-borne pathogens, such as salmonella and E-coli, and viruses such as flu, norovirus and hepatitis A.
"My PhD focused on sewer design to reduce concrete corrosion, however I graduated right around the time of COVID-19 and saw an opportunity to monitor and study the pandemic," said Dr Li.
"I'm grateful for the opportunity to receive the WH Gladstones award, particularly as an early career researcher, to explore the potential to create early-warning systems for COVID-19 and other diseases. I hope this work can benefit the community and inspire other women in science."
