Aston University’s artificial intelligence expertise has helped health agencies to estimate daily transmission rates of infections such as COVID-19 within the U.K. The model makes use of antibody data collected at blood donation centres.
The data obtained has allowed academics to estimate the proportion of people who were going undiagnosed. The approach has a considerable advantage over current epidemiological models. These tend not to be less effective at estimating hidden variables such as daily infection rates.
The Birmingham based university collaborated with researchers at the Universidade Federal de Minas Gerais in Brazil to develop the model. The algorithm was based upon a large longitudinal study into data obtained from Brazilian blood donor centres (7,837 blood donors in seven cities in Minas Gerais, Brazil during March–December 2020).
The approach was to use a compartmental model, which is a general modelling technique often applied to the mathematical modelling of infectious diseases. This approach enables researchers to assess and predict things such as how a disease spreads, the total number infected, the duration of an epidemic. In addition, it is possible to estimate various epidemiological parameters such as the pathogen reproductive number.
The study looked at the reported number of SARS-CoV-2 cases along with serology results (diagnostic methods to identify antibodies and antigens in patients’ samples). This revealed hidden variables like daily values of transmission rates and cumulative incidence rates of reported and unreported cases.
The model also provided an insight into changes in the infection rate (or how many people each case infected on average). Other models led to the assumption this was a fixed value over a long duration of time. The AI model, in contrast, showed the dynamics of the spread of COVID-19 alter far faster than realized.
The data was regarded as especially important for understanding the proportion of people who were going undiagnosed. This included including for asymptomatic and mildly symptomatic people.
The research paper appears in Emerging Infectious Diseases as it is titled “SARS-CoV-2 IgG Seroprevalence among Blood Donors as a Monitor of the COVID-19 Epidemic, Brazil”. This gave the scientists the ability to adopt a more refined view of the infection rates and relative rate of immunity compared to official measurements.
Several aspects of the model are regarded as important, especially during the early days of a pandemic and when applied to similar viral diseases.
Going forwards, the researchers aim to strengthen the accuracy of the model by introducing changes to account for vaccination effects, waning immunity and the potential emergence of new variants.
