How Big Data and Artificial Intelligence Can Help Better Manage the COVID-19 Pandemic

Abstract

SARS-CoV2 is a novel coronavirus, responsible for the COVID-19 pandemic declared by the World Health Organization. Thanks to the latest advancements in the field of molecular and computational techniques and information and communication technologies (ICTs), artificial intelligence (AI) and Big Data can help in handling the huge, unprecedented amount of data derived from public health surveillance, real-time epidemic outbreaks monitoring, trend now-casting/forecasting, regular situation briefing and updating from governmental institutions and organisms, and health facility utilization information. The present review is aimed at overviewing the potential applications of AI and Big Data in the global effort to manage the pandemic.

Artificial Intelligence-based Disease Surveillance Amid COVID-19 and Beyond: A Systematic Review Protocol

“…However, not all hope is lost. One of the most promising tools that can help health experts better curb and control the spread of COVID-19 is AIbased disease surveillance systems [44][45][46][47][48].…”

“…Adding the fact that global health crises can often cause economic fallout and resources constraints, the cost-effective nature of technological solutions increases their potential as candidate solutions for solving various healthcare issues [37,[41][42][43]. Lastly, and perhaps most importantly, the potential for some technological-based solutions, such as arti cial intelligence (AI) powered disease surveillance systems, in identifying infection cases with high accuracy, may be the most desirable quality governments need to effectively control the spread of diseases [44][45][46][47][48][49][50]. AI can be understood as "the theory and development of computer systems able to perform tasks normally requiring human intelligence, such as visual perception ... and decision-making" [51].…”

“…One key facilitating factor that further popularized AI-based disease surveillance is the availability of large datasets, mainly coming from (1) data actively generated by netizens (e.g., social media user generated data, news reports, etc. ), (2) data collected about the individuals' passive human-computer interaction (e.g., background data collected by apps and electronic devices), and (3) data collated independent of individuals' human-computer interactions (e.g., video surveillance footage) [48,[53][54][55]. BlueDot, the data analytics company that rst identi ed and reported an unusual cluster of pneumonia cases (i.e., the COVID-19 outbreak) eight days before the World Health Organization issued a global warning about a potential global crisis in the making [56], adopted data actively generated by netizens (e.g., social media posts, newsfeeds) and data collated independent of individuals' human-computer interactions (e.g., information obtained from airline ticketing systems) to feed into their AI systems to generate insights [57].…”

“…In previous work, AI-systems that synthesize multiple measures such as respiration rate, heart rate, and facial temperature, have been successfully applied to in uenza surveillance with high detection accuracy [49]. In other words, AI-based disease surveillance can accomplish, essentially, the ability to swiftly identify individuals who have been infected with or exposed to infectious agents, a challenge that remains to be one of the deadliest problems that health experts face in the ght against COVID-19 [44][45][46][47][48]. Lacking the ability to discern potential COVID-19 infected individuals from the rest of the population is also a key reason why governments chose to lockdown entire populations in early days of COVID-19 outbreak [58][59][60], with signi cant economic cost.…”

Data Analytics for Predicting COVID-19 Cases in Top Affected Countries: Observations and Recommendations

“…Pinter et al [ 20 ] have developed a hybrid machine learning approach to forecast COVID-19 cases in Hungary. For extensive study and more details about the forecasting approaches for COVID-19, the interested readers are referred to the work by Bragazzi et al [ 21 ] who review the potentials of applying artificial intelligent and big data based approaches in predicting and managing the COVID-19 Pandemic outbreak. The pitfall of the previous studies is that they have focused on historical data of confirmed COVID-19 cases, while some studies have considered some factors like temperature and patient sex.…”