Next-Generation In Vitro Pulmonary Platforms for Respiratory Disease Modelling and Therapeutic Development: Current Advances and Future Prospects.

Pulmonary diseases such as Chronic obstructive pulmonary disease (COPD), asthma, pulmonary fibrosis, and acute respiratory infections remain a major global health challenge due to their complex pathophysiology and limited therapeutic options. Conventional 2D cultures and animal models have provided foundational insights; however, they often fail to accurately replicate the human lung's intricate architecture, immune interactions, and patient-specific variability. Recent advances in vitro technologies have transformed pulmonary research, enabling the generation of physiologically relevant and translational disease models. The review highlights the progression of lung research platforms from traditional monolayer cultures to advanced systems such as air-liquid interface models and 3D lung organoids. These cutting-edge models more effectively mimic the biochemical, mechanical, and spatial microenvironment of the respiratory system, enhancing the fidelity of disease modelling and drug screening. In parallel, the integration of computational modelling and artificial intelligence (AI) has emerged as a powerful synergistic approach. AI-driven analytics facilitate high-throughput imaging, biomarker discovery, and patient-stratified therapeutic prediction, while computational tools simulate disease networks, mechanobiological interactions, and pharmacological responses. The convergence of these technologies supports a deeper understanding of pulmonary disease progression and accelerates the development of precision therapeutics. Collectively, this review underscores the transformative potential of combining in vitro lung models with advanced computational and AI methodologies. This synergy not only improves translational relevance and reduces reliance on animal testing but also paves the way for personalised interventions that better address the complexity of human pulmonary disease.