Exploring Healthcare and Public Health Implications via Temporal Analysis of Human–Virus Interactions Leveraging Adversarial Networks and Saliency Maps
Viruses are prone to rapid mutations that enhance traits like transmissibility and resistance to treatments. The adaptability complicates the development of effective vaccines. The study investigates the dynamic interactions between humans and COVID-19 virus through time-series analysis, focusing on...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | Applied Artificial Intelligence |
| Online Access: | https://www.tandfonline.com/doi/10.1080/08839514.2025.2476237 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Viruses are prone to rapid mutations that enhance traits like transmissibility and resistance to treatments. The adaptability complicates the development of effective vaccines. The study investigates the dynamic interactions between humans and COVID-19 virus through time-series analysis, focusing on how these interactions evolve over time. A novel approach was introduced using adversarial networks to model the relationship between human and viral entities, capturing their temporal dynamics. Two interdependent models are trained within a framework, allowing us to forecast interactions and identify influential features. Saliency maps were utilized to visualize the factors affecting these interactions over time. The technique helps reveal how specific viral properties shift in response to human factors. The findings of the study aim to enhance the understanding of human-virus dynamics, particularly for COVID-19, offering potential insights for public health strategies and interventions. By combining adversarial networks with saliency visualization, the study provides an intensive study for analyzing and interpreting complex temporal interactions between humans and viruses. |
|---|---|
| ISSN: | 0883-9514 1087-6545 |