AI in Combating the COVID-19 Pandemic

Abstract

The SARS-CoV-2 virus, the COVID-19 disease, and the resulting pandemic have reshaped the entire world in an unprecedented manner. Massive efforts have been made by AI communities to combat the pandemic. What roles has AI played in tackling COVID-19? How has AI performed in the battle against COVID-19? Where are the gaps and opportunities? What lessons can we learn to enhance the ability of AI to battle future pandemics? These questions, despite being fundamental, are yet to be answered in full or systematically. They need to be addressed by AI communities as a priority despite the easing of the omicron infectiousness and threat. This article reviews these issues with reflections on global AI research and the literature on tackling COVID-19. It is envisaged that the demand and priority of developing “pandemic AI” will increase over time, with smart global epidemic early warning systems to be developed by a global collaborative AI effort.

COVID-19 Modeling: A Review

“…COVID-19 pandemic is a complex physical-social-technical system, which further involves the cyberspace as well [43]. Accordingly, COVID-19 modeling has to address their general and domain-specific complexities, challenges and research questions in the continuum, take advantage of the advances in their related physical, social, technical and cyber sciences, and involve their domain-specific knowledge and factors.…”

“…The review in [56] lists and categorizes the applications of various AI techniques in COVID-19 detection, diagnosis, virology, pathogenesis, drug and vaccine development, and coronavirus transmission and prediction. The review in [43] provides a brief summary of the challenges, global AI research, the modeling trends and gaps in combating COVID-19. The report provided by AAAS [1] provides a brief summary of selective applications and the impact of AI in fighting COVID-19 in particular in the US.…”

How Have Global Scientists Responded to Tackling COVID-19?

“…Third, we aim to understand how various modeling techniques (Cao and Liu 2021), including applied mathematics, AI and data science (including shallow and deep learning and analytics) (Cao 2022), epidemic modeling, and other methods such as simulation, have contributed to quantifying COVID-19 in terms of their trending effects. We thus explore the following questions:…”

“…In particular, as quantifying COVID-19 is essential for understanding and managing the pandemic, it is important to understand what modeling techniques including in mathematics, epidemiology, AI, data science, machine learning, deep learning, and social science etc. have been applied and where they have played their part in modeling COVID-193 (Cao and Liu 2021; Cao 2022).…”

“…Second, to quantify how G20 and OECD countries and regions have performed in tackling COVID-19, we explore the indication to the following questions: What are the publication number and impact of G20 countries and regions? What is the disciplinary publication impact of G20 countries and regions? What is the publication-averaged mean publication impact of G20 countries and regions? What are the publication number and mean publication impact of OECD countries and regions? Third, we aim to understand how various modeling techniques (Cao and Liu 2021), including applied mathematics, AI and data science (including shallow and deep learning and analytics) (Cao 2022), epidemic modeling, and other methods such as simulation, have contributed to quantifying COVID-19 in terms of their trending effects. We thus explore the following questions: What are the main keywords concerned in modeling COVID-19 globally? What are the main modeling keywords concerned in different disciplines? What are the main keywords concerned in modeling COVID-19 in the two major economic and technical powers USA and China respectively? What are the top-50 concerned problems and their publication impact? What are the top-50 concerned problems and their modeling techniques? What are the top-50 modeling techniques and their publication impact? What are top-10 monthly concerned problems in modeling COVID-19? What are the top-10 problems concerned in the top-10 published countries? What are the top-10 modeling techniques applied by the top-10 published countries? What are the top-10 problems and techniques concerned by the top-10 published countries? Lastly, we aim to understand the relations between the COVID-19 research outcomes and a country’s coronavirus infections, deaths, and their economy by exploring the following questions: How does global first-authored countries’ GDP per capita correlate with their publications and cumulative publication impact? How does a G20 country’s GDP correlate with their publication quantity and impact? How does an OECD country’s GDP correlate with their publication quantity and impact? How does global monthly publications correlate with the global monthly infections and deaths in major disciplines? How does a G20 country’s coronavirus infections and deaths correlate with their publications? How does an OECD country’s coronavirus infections and deaths correlate with their publications? …”