After the first year of the pandemic in 2021, a partnership was established between the University of Peradeniya in Malaysia and the Institute of Policy Studies from Sri Lanka, for the project entitled "An Artificial Intelligence Framework for Threat Assessment and Containment for COVID-19 and Future Epidemics while Mitigating the Socioeconomic Impact on Women, Children, and Underprivileged Groups''. The project aims to develop AI to conduct contact tracing and socioeconomic impact mitigation actions in a more informed, socially conscious, and responsible manner in case of the next wave of COVID-19 infections or a different future infectious disease. This project consists of two groups of academics from Sri Lanka and Malaysia. The University of Peradeniya and the Institute of Policy Studies from Sri Lanka are collaborating with three universities in Malaysia namely the University Tenaga Nasional, Manipal University College, and Multimedia University are leading this study. The following are a few key takeaways about the grantee experience from the Sri Lankan research team about being part of the AI4COVID project.
Research Methodology
Artificial Intelligence (AI) and Data Science is a new branch of study that deals with scientific methodologies, processes, and techniques drawn from different domains to extract knowledge from structured and unstructured data. AI4COVID is carried out under a number of work packages (WPs) designed to address different specific objectives of the project. There are three WPs and each one is divided into tasks (Ts). WP1 is about data collection for the project by means of websites, surveys, and questionnaires. The collected data is analysed under WP2 using AI techniques to generate useful information for disease containment and policy making. The outcomes of the project will be bridged into policymaking by conveying them to the governing bodies and health sector of the country.
Figure 1. Workflow of the Sri Lanka AI4COVID project
The AI4COVID project gave our research team the opportunity to collaborate with experts from a variety of fields including engineering, sociology, and public policy. Furthermore, the project helped our research team expand their thinking and broaden their skill sets. When asked about his experience working with the project, a former research team member expressed:
“Participating in the AI4COVID project was a career-defining moment for me. Working on the project exposed me to different perspectives, approaches, and methodologies and I was able to integrate these into my own work. It challenged me to think beyond the boundaries of my field of work, and has had a lasting impact on my research direction.”
Key Successes of the AI4COVID
The project has successfully collected a large data set from household surveys (11,532 individual data points) and qualitative and mobility-related data from the Focus Group Discussions (FGDs) and in-depth interviews (18 FGDs and 189 in-depth interviews). Our research team hopes to gather more tracking data from a mobile app in the upcoming months. Moreover, the research team is in the process of developing a simulator that can be tuned through the findings of the aforementioned surveys and interviews, so it would be capable of reflecting upon the actual dynamics of people. This simulator can be used to identify the impact of social distancing, different vaccinations, and containment strategies. The aim of the simulator is to understand how an individual’s action and behavior at the micro level affects the pandemic at the macro level.
Photo 1. Focus Group Discussion (FGD) Session—Photo by AI4COVID Team Sri Lanka
A mid-term review of the AI4COVID project was successfully held in Malaysia in December 2022 with the participation of the research teams from Sri Lanka and Malaysia presenting the current progress made by both countries and to discuss potential collaborative efforts in the future.
Photo 2. Mid-Term Review Meeting Held in Malaysia—Photo by AI4COVID Team Malaysia
Final Outputs and Key Lessons Learned from the AI4COVID
Currently, the research team is working on utilizing collected data to achieve its goals. For instance, the AI team of the research group has started to tune up the simulator (motion model) using the collected data on mobility. Moreover, the social sciences team is conducting further research to analyze the impact of the COVID-19 pandemic on both the education and mobility of people. Moving forward, the research team plans to broaden their investigation by examining the pandemic's effect on the income of individuals. Additionally, workshops are scheduled to be conducted in the coming months to inform medical officers and policymakers on the significance of these ongoing studies for the advancement of healthcare systems and development of the medical and economic policies in the country in order to be better prepared when facing similar situations in the future. Although the project mainly focuses on COVID-19, the methodology and the designs can be readily applied to other infectious diseases as well.
The AI4COVID project is being implemented with grant support from Canada’s International Development Research Centre (IDCR) and the Swedish International Development Cooperation Agency (Sida) as part of the Global South AI4COVID Program.
To learn more about our work visit our project website: ai4covid.lk
