Using AI and data against the pandemic: New projects in Global South
The Global South AI4COVID Response Program (originally known as the COVID-19 Global South AI and Data Innovation Program) is co-funded by IDRC and the Swedish International Development Cooperation Agency (Sida). The nine selected projects will be funded over a 24-month period and cover a diverse range of issues across Asia, Latin America, and sub-Saharan Africa.
The projects will mobilize multidisciplinary consortia who will use AI and data science to combat the negative impacts of COVID-19 and strengthen health systems to improve pandemic response. Overall, the initiative aims to deepen the understanding of governments, AI practitioners, and the public on how to develop and scale responsible and evidence-based AI and data science approaches that support COVID-19 response and recovery in developing countries.
The initiative is particularly concerned with ensuring that these projects are gender responsive and culturally appropriate, community specific, and based on local needs and contexts. It also aims to inform policies that support and build trust in AI and data science responses to epidemics and mitigate potential harms. Finally, it will aim to strengthen the capacity of health systems in developing countries to respond to epidemics using AI and data science techniques.
“These projects should inform policies that support and build trust in responsible, inclusive, artificial intelligence and data responses to COVID-19,” says IDRC President Jean Lebel. “We also anticipate that they will strengthen health systems in developing countries and improve the ability to respond to all kinds of epidemics in the future, using AI and data science techniques.”
The response to the funding call in June received 314 submissions, of which 153 eligible proposals were evaluated by an independent panel of experts. The panel shortlisted 20 proposals and IDRC and Sida selected the final nine projects based on overall scores and the thematic and geographic distribution of the projects. To ensure that these projects are ultimately able to inform policies at all levels and build trust in AI and data science responses, they were also selected for their gender responsiveness, cultural sensitivity, community focus, and local considerations.
"Eight months into the global pandemic, COVID-19 continues to affect lives right around the world", says AnnaMaria Oltorp, head of Sida’s Research Cooperation Unit. "Sida is pleased to collaborate with IDRC on this important opportunity for technology and data researchers in low- and middle-income countries to help address the current challenges nimbly and to stay on top of future outbreaks.”
Overview of the nine projects:
- Organization: Universidad de los Andes
- Country of impact: Colombia
- Vulnerable communities in Colombia are affected not only by the coronavirus disease, but also by disinformation about the pandemic. Universidad de los Andes will develop Al models to shed light on the risk of disease progression to vulnerable communities and its impact on other diseases. They will also address the “infodemic” by analyzing popular social networks, users who act as ”super spreaders” of information and misinformation, and common reactions to policies and events of social importance.
- Organization: Social Science Academy of Nigeria
- Country of impact: Nigeria
- The project will leverage data science and AI tools to develop a better understanding of how COVID-19 aggravates gender-based violence in Nigeria and how to support women who are at the greatest risk.
- Organization: Centro Interdisciplinario de Estudios en Ciencia, Tecnología e Innovación (CIECTI)
- Country of impact: Argentina
- CIECTI will leverage electronic health records to pilot new models to support early COVID-19 detection in communities, improve data collection in vulnerable communities, and work with the National Ministry of Health to contain the disease.
- Organization: York University
- Countries of impact: Botswana, Cameroon, Mozambique, Namibia, Nigeria, Rwanda, South Africa, Zimbabwe
- York University has joined forces with epidemiologists, modelers, physicists, statisticians, software engineers, and data scientists across Africa to integrate the power of predictive modelling and simulations with the capacity of a comprehensive COVID-19 monitoring dashboard that will be used to predict epidemic trends and inform decision-making and real-time management across Africa.
- Organization: Université Cheikh Anta Diop
- Countries of impact: Mali, Senegal
- While Senegal and Mali are eager to use AI methods to improve contact tracing efforts, the countries contend with a number of social and political challenges. This research will support epidemiological modelling of COVID-19 in Senegal and Mali while addressing the social and political obstacles in connection with the use of AI technologies and health control measures within a framework of ethics and human rights.
- Organization: University of Peradeniya
- Countries of impact: Malaysia, Sri Lanka
- These new modelling efforts will focus on supporting Malaysian and Sri Lankan efforts to map the COVID-19 crisis and mitigate COVID-19's economic impacts on women, children, and underprivileged groups.
- Organization: University of Rwanda
- Country of impact: Rwanda
- Government and other public health institutions are calling for initiatives that can support the increased availability of data to support evidence-based policymaking. However, much of the available data is fragmented, incomplete, and scattered across multiple institutions such as clinics, hospitals, and testing sites. This project will innovate to harmonize data to understand and predict the impact of COVID-19 and other infectious diseases. It will also focus on new innovations to make best use of the data to advance policy and care decisions.
- Consortia lead: Makerere University
- Country of impact: Uganda
- This project aims to deliver a set of contextualized and equitable end-to-end AI and data systems that target the surveillance and management of COVID-19 and future pandemics that may affect Uganda. The project will focus on using AI systems to analyze radio conversations to improve understanding of and to address public perceptions of COVID-19 and interventions aimed at understanding their connection with COVID-19 transmissions and interventions. It will also explore new innovations for diagnosing and treating COVID-19 using AI tools to support screening.
- Consortia lead: APHRC
- Countries of impact: Kenya and Malawi
- This project will harmonize data to compare and understand the impacts of different public health decisions taken in Malawi and Kenya. It seeks to understand COVID-19 transition dynamics and their impact on health, education, work, and transport, including how interventions work and where they work best.
- Organization: African Population and Health Research Center (APHRC), Kenya
- Countries of impact: Kenya, Malawi
- This project proposes to develop the key elements of a coordinated pan-African COVID-19 data ecosystem. We will build a robust suite of data standards and technologies, diverse data integration methodologies, using the power of AI and data science for analysis and oversight through a trusted governance and policy environment. The harmonized data represents a global public good which will deliver insights for evidence-based answers on key questions about COVID-19.
- Organization: University of Makerere, Uganda
- Country of Impact: Uganda
- In sub-Saharan Africa data imbalances and underrepresentation can easily arise due to unequal access to government and private services where data is collected due to socio-demographic conditions. COAST will address these challenges through three specific objectives: 1- To strengthen data systems resulting in usable and equitable datasets for AI-driven COVID-19 responses and future pandemics. 2- To develop and deploy AI-driven detection and diagnosis tools for improved patient care and management. 3- To model and evaluate COVID-19 interventions for targeted government responses based on the fused datasets from objectives 1 and 2.