Artificial intelligence may hold the key to limiting the spread of infection in future pandemics. In a new study, researchers at the University of Gothenburg investigated how machine learning can be used to find effective testing methods during epidemic outbreaks, allowing for better outbreak control.
The researchers developed a method in the study to improve testing strategies during epidemic outbreaks and predict which individuals offer the best potential for testing with relatively limited information. The researchers discovered that when the method is used, the outbreak can be quickly brought under control, whereas random testing leads to an uncontrolled spread of the outbreak with many more infected individuals.
“This can be a first step towards society gaining better control of future major outbreaks and reduce the need to shutdown society,” says Laura Natali, a doctoral student in physics at the University of Gothenburg and the lead author of the published study.
Machine Learning (ML) can be used to find effective testing methods during epidemic outbreaks, thereby helping to better control the outbreaks, a new study said. In a new study, researchers have investigated how machine learning can be used to find effective testing methods during epidemic outbreaks, thereby helping to better control the outbreaks.
ML is a type of Artificial Intelligence that can be defined as a mathematical model in which computers are trained to see connections and solve problems using various data sets. In a simulation of a pandemic outbreak, the research team used machine learning to estimate infections in the rest of the population based on data from the first confirmed cases. Data from the infected person’s network of contacts and other information, such as who they have been in close contact with, where they have been in close contact with them, and how long they have been in close contact with them, were used.
Machine learning is a subset of artificial intelligence that can be defined as a mathematical model in which computers are taught to see connections and solve problems using various data sets. In a simulation of an epidemic outbreak, the researchers used machine learning to estimate infections in the rest of the population based on information from the first confirmed cases. Data from the infected person’s network of contacts and other information were used: who they have been in close contact with, where they have been in close contact with them, and how long they have been in close contact with them.
“In the study, when the method is used, the outbreak can be quickly brought under control, whereas random testing leads to an uncontrolled spread of the outbreak with many more infected individuals. Under real-world conditions, information such as demographic data, age, and health-related conditions can be added to improve the method’s effectiveness even further. If immunity after the disease is only temporary, the same method can be used to prevent reinfections in the population.”
She emphasizes that the study is a simulation and that real-world testing is required to improve the method even further. As a result, it is too early to apply it to the ongoing coronavirus pandemic. At the same time, she sees the study as a first step toward being able to implement more targeted initiatives to reduce the spread of infections, because the machine learning-based testing strategy automatically adapts to disease-specific characteristics. She cites the possibility of easily predicting whether a specific age group should be tested or whether a specific geographic area, such as a school, community, or neighborhood, is a risk zone.
“When a large outbreak begins, it is critical to quickly and accurately identify infectious individuals. There is a significant risk of failing to achieve this with random testing, but with a more goal-oriented testing strategy, we can find more infected individuals and thus gain the necessary information to reduce the spread of infection. We demonstrate how machine learning can be used to create this type of testing strategy “she claims.
There have been few previous studies that have looked into how machine learning can be used in pandemic situations, particularly with a clear focus on finding the best testing strategies. “We demonstrate that it is possible to predict who would be most beneficial to test using relatively simple and limited information. This allows for more efficient use of testing resources.”
This study is one of the first to investigate how machine learning can be used to mitigate pandemics, and it is one of only a few that has a clear focus on determining the best testing strategy.
