Artificial intelligence in identifying COVID-19 patients based on inflammatory parameters
11th Mediterranean Conference on Embedded Computing, MECO 2022
; 2022.
Article
in English
| Scopus | ID: covidwho-1948828
ABSTRACT
The virus SARS-Co V -2 that has caused a pandemic of COVID-19 in 2019 is still a major concern for health care systems. The reason for this is the fact that the outcome of the disease is difficult to predict, as deadly complications can occur in all people. Diagnosing COVID-19 relies on polymerase chain reaction (PCR) testing and antigen testing, both of which require special referral. The aim of this study was to develop artificial intelligence (AI) expert system which will facilitate COVID-19 diagnosis based on parameters that can be readily collected from blood specimens. The database contains 1000 samples, divided into 2 categories (1) healthy and (2) sick subjects The following parameters were used CRP, LDH, SE, AST, ALT, D-dimer and IL-6. The sensitivity of the developed system was 100%, specificity 98.33%, and accuracy 99.67%, on the basis of which we can conclude that the use of AI in the diagnosis of COVID19 has a significant potential. © 2022 IEEE.
Full text:
Available
Collection:
Databases of international organizations
Database:
Scopus
Language:
English
Journal:
11th Mediterranean Conference on Embedded Computing, MECO 2022
Year:
2022
Document Type:
Article
Similar
MEDLINE
...
LILACS
LIS