Berkeley Lab using AI to estimate Covid-19 seasonal cycle

20 May 2020 (Last Updated May 20th, 2020 17:04)

Scientists at Lawrence Berkeley National Laboratory (Berkeley Lab) are using machine learning to assess whether Covid-19 has a seasonal aspect that will lead it to wane in summer and resurge in winter.

The computing work will be conducted at the National Energy Research Scientific Computing Center (NERSC), a US Department of Energy Office of Science user facility located at Berkeley Lab. A range of health and environmental datasets are being used, alongside high-resolution climate models and seasonal forecasts.

The research team will take advantage of abundant health data on the severity, distribution and duration of the Covid-19 outbreak in different countries, as well as various public health interventions alongside demographics, climate, population mobility dynamics and weather factors. The initial goal of the research is to predict environmental factors that may impact disease transmission in each US county by looking at the geographical differences in disease behaviour that have already been reported.

Berkeley Lab computational biologist Ben Brown says: “There are cities where [Covid-19] behaves as if it's the most infectious disease in recorded history. Then there are cities where it behaves more like influenza. It is really critical to understand why we see those massive differences.”