Introduction. This review aims to investigate modern methods of applying
artificial intelligence to diagnose SARS Cov-2 and predict the development
of potential emergencies. Methods. The most commonly used electronic
databases, such as Scopus and Medline during 2020, were searched. A
narrative approach was used to synthesize the extracted data. Results. In
this review paper, it has been shown that the application of artificial
intelligence plays a significant role in virus diagnosis and prognosis in
clinical trials. It allows resources to be used much more rationally, such
as respirators, in hospitals, during the treatment of SARS Cov-2 and the
prediction of possible mortality. The obtained results are from the analysis
performed on 120 papers and studies that were electronically taken from
papers published on Scopus and Pub Med line. Most commonly used artificial
intelligence techniques are convolutional neural networks and machine
learning. Conclusions. Included studies showed that artificial intelligence
can significantly improve the treatment of SARS Cov-2, although many of the
proposed methods have not yet been clinically accepted. In addition, more
effort is needed to develop standardized reporting protocols or guidelines
on applying artificial intelligence into conventional clinical practice.
This technology is suitable for fast and accurate diagnosis, prediction and
monitoring of current patients and prognosis of disease development in
future patients.