Potential of artificial intelligence to accelerate diagnosis and drug discovery for COVID-19

• Main Authors: Indira Mikkili, Abraham Peele Karlapudi, T. C. Venkateswarulu, Vidya Prabhakar Kodali, Deepika Sri Singh Macamdas, Krupanidhi Sreerama
• Format: Article
• Language: English
• Published: PeerJ Inc. 2021-10-01
• Series: PeerJ
• Subjects: Machine learning; Neural networks; Drug discovery; Reverse transcriptase polymerase chain reaction; Artificial intelligence; Computed tomography
• Online Access: https://peerj.com/articles/12073.pdf

The coronavirus disease (COVID-19) pandemic has caused havoc worldwide. The tests currently used to diagnose COVID-19 are based on real time reverse transcription polymerase chain reaction (RT-PCR), computed tomography medical imaging techniques and immunoassays. It takes 2 days to obtain results from the RT-PCR test and also shortage of test kits creating a requirement for alternate and rapid methods to accurately diagnose COVID-19. Application of artificial intelligence technologies such as the Internet of Things, machine learning tools and big data analysis to COVID-19 diagnosis could yield rapid and accurate results. The neural networks and machine learning tools can also be used to develop potential drug molecules. Pharmaceutical companies face challenges linked to the costs of drug molecules, research and development efforts, reduced efficiency of drugs, safety concerns and the conduct of clinical trials. In this review, relevant features of artificial intelligence and their potential applications in COVID-19 diagnosis and drug development are highlighted.