Artificial Intelligence for Predicting Lung Immune Responses to Viral Infections: From Mechanistic Insights to Clinical Applications

Artificial intelligence (AI) is increasingly transforming biomedical research and patient care by integrating complex biological, radiological, and healthcare information. In the field of viral respiratory infections, AI-driven approaches have shown great promise in elucidating the complexity of lung immune responses and the dynamic interplay between host and pathogen. Applications include predicting cytokine storm and acute respiratory distress syndrome (ARDS), integrating imaging findings with immunological and laboratory data, and identifying molecular and cellular signatures through single-cell and multi-omics analyses. Similar methodologies have been applied to influenza and respiratory syncytial virus (RSV), providing insights into the mechanisms distinguishing protective from maladaptive pulmonary immunity. This narrative review summarizes current evidence on how AI can evolve into a form of translational intelligence, capable of bridging mechanistic immunology with clinical application. The review explores AI-based models for disease severity prediction, patient stratification, and therapeutic response assessment, as well as emerging approaches in drug repurposing and vaccine response prediction. By integrating biological complexity with clinical context, AI offers new opportunities to uncover immune signatures predictive of antiviral or immunomodulatory efficacy and to guide personalized management strategies.