Shaping the future of pandemic defense: A review of breakthrough COVID-19 detection techniques.

The COVID-19 pandemic has prompted an urgent need for reliable and efficient diagnostic tools to combat its spread. Since its origin in 2019, researchers have been trying to find out effective methods for identifying and classifying COVID-19 from other similar disease like pneumonia. This review paper gives a comparable overview of various diagnostic techniques employed by various experts. Artificial Intelligence (AI) based methods, like Machine Learning (ML) and Deep learning (DL) have widely been used for predicting COVID-19. These models usually generate good accuracy rates but still there is a scope of improvement. Hence, attention was shifted towards hybrid approaches which show better detection accuracy results then individual AI models. However, both AI and hybrid models overlook human social connections and migration patterns, which are diverse and not well-mixed in populations. COVID-19 modeling should consider the heterogeneity of individuals, especially as it disproportionately affects the elderly and those with underlying medical conditions like cancer, heart disease, diabetes, and hypertension. Furthermore, people's mobility patterns and behaviors significantly influence virus spread dynamics. Therefore, role of multi-agent systems (MAS) comes into play, wherein each agent represents different data or information. However, not much work has been done in this filed and hence it needs to be explored more in future studies. Furthermore, it was noted that a significant proportion of researchers favored the utilization of CNN and its variants and SVM classifiers, while a minority opted for RF and regression models for the purpose of classification. Overall, this review provides insights into the strengths and limitations of each approach, emphasizing the need for a comprehensive strategy that harnesses multi-agent systems for COVID-19 detection. [ABSTRACT FROM AUTHOR]