Deep Learning based Detection and Prediction of Omicron Diagnosis on Collected Symptoms

ABSTRACT

Active inspection of Omicron variant enables rapid and effective analysis of COVID-19 variant. It can diminish the encumbrance on health systems. Detection and forecast models combine many qualities for calculation. The hazards of infection have progressed. The goal is to aid health care professionals globally in case of critical Omicron long-suffering persons, specifically in the situation of inadequate well-being means. The developed deep learning research approach was trained on the dataset of 900 verified persons (among which 500 were inveterate, to partake Omicron variant). The training test-set delimited the dataset from the succeeding days (700 verified persons, of which 300 were inveterate, to partake Omicron). The proposed research prototype spotted and forecasted Omicron examined outcomes with good correctness and little error rate using only six unique features such as gender, age = 50 ages, known traces with a disease distinct, and the presence of four preliminary medical indications. The developed research prototype notices Omicron infected cases via easy qualities retrieved through posing elementary queries or data queries. The proposed research outline can be useful, amid additional attention, to arrange to examine Omicron while checking properties are insufficient. © 2022 IEEE.