[Paper Review] A Survey on Applications of Artificial Intelligence in Fighting Against COVID-19

Jianguo Chen, Kenli Li+3 more|arXiv (Cornell University)|Jul 4, 2020

COVID-19 diagnosis using AI211 references22 citations

TL;DR

This survey comprehensively reviews the application of artificial intelligence (AI) in combating COVID-19, focusing on disease detection, virology, drug and vaccine development, and epidemic prediction. It synthesizes AI techniques—particularly in medical imaging, genomics, and predictive modeling—demonstrating their critical role in accelerating diagnosis, therapy, and public health interventions during the pandemic.

ABSTRACT

The COVID-19 pandemic caused by the SARS-CoV-2 virus has spread rapidly worldwide, leading to a global outbreak. Most governments, enterprises, and scientific research institutions are participating in the COVID-19 struggle to curb the spread of the pandemic. As a powerful tool against COVID-19, artificial intelligence (AI) technologies are widely used in combating this pandemic. In this survey, we investigate the main scope and contributions of AI in combating COVID-19 from the aspects of disease detection and diagnosis, virology and pathogenesis, drug and vaccine development, and epidemic and transmission prediction. In addition, we summarize the available data and resources that can be used for AI-based COVID-19 research. Finally, the main challenges and potential directions of AI in fighting against COVID-19 are discussed. Currently, AI mainly focuses on medical image inspection, genomics, drug development, and transmission prediction, and thus AI still has great potential in this field. This survey presents medical and AI researchers with a comprehensive view of the existing and potential applications of AI technology in combating COVID-19 with the goal of inspiring researchers to continue to maximize the advantages of AI and big data to fight COVID-19.

Motivation & Objective

• To provide a comprehensive overview of AI applications in fighting COVID-19 across multiple domains.
• To identify key AI techniques used in disease detection, virology, drug development, and transmission prediction.
• To summarize publicly available data and resources for AI-based COVID-19 research.
• To highlight major challenges and propose 11 future research directions for AI in pandemic response.

Proposed method

• Systematic review of peer-reviewed literature and technical reports on AI applications in COVID-19 research.
• Categorization of AI applications into four core areas: disease detection and diagnosis, virology and pathogenesis, drug and vaccine development, and epidemic prediction.
• Analysis of AI techniques including deep learning for medical image analysis (e.g., CXR and CT), transfer learning for imbalanced data, and NLP for fake news detection.
• Integration of AI with genomics and structural biology to predict viral protein functions and drug-protein interactions.
• Use of machine learning models to analyze large-scale epidemiological and social media data for transmission forecasting.
• Construction of knowledge graphs and social network models for patient contact tracing and risk assessment.

Experimental results

Research questions

• RQ1 How has AI been applied in the detection and diagnosis of COVID-19 using medical imaging and clinical data?
• RQ2 What role has AI played in understanding SARS-CoV-2 virology, including viral structure and host interactions?
• RQ3 How have AI-driven approaches accelerated drug and vaccine development during the pandemic?
• RQ4 In what ways has AI contributed to predicting and modeling the transmission and spread of the virus?
• RQ5 What are the key data resources and technical challenges enabling or limiting AI applications in COVID-19 research?

Key findings

• AI has significantly accelerated non-invasive diagnosis through automated analysis of chest X-ray and CT scans, reducing reliance on nucleic acid testing.
• Deep learning models achieved high accuracy in detecting COVID-19 from medical images, with some studies reporting AUC values above 0.90.
• AI-enabled genomics analysis helped identify the zoonotic origin of SARS-CoV-2 and revealed structural similarities with bat and MERS-related coronaviruses.
• Transfer learning and few-shot learning techniques improved model performance on small, imbalanced datasets of positive and negative COVID-19 cases.
• AI-driven epidemic models successfully predicted outbreak timing, transmission routes, and regional spread using mobility and case data.
• AI-based systems were deployed for remote diagnosis, fake news filtering, and intelligent contact tracing, enhancing public health response efficiency.