Modeling of Pneumonia and Acute Lung Injury: Bioinformatics, Systems Medicine, and Artificial Intelligence

Abstract Pneumonia and acute lung injury cause the highest mortality of all infectious disease, and tremendous medical challenges, e.g. due to newly emerging pathogens and increasing rates of antimicrobial resistance. Increasingly performant computer systems have been used for over a decade to model complex biomedical systems, thereby establishing a field of research that is closely intertwined with conventional biomedical diagnostics and basic research. The application of mathematical modeling and bioinformatics to unravel the patho-mechanisms of the diverse forms of pneumonia ranges from epidemiological modeling of pneumonia and deep learning to systems biology and bioinformatics. While the first two focus on predicting population-wide disease progression and establishment of patterns in data that inform about predictive value, the latter pair aims to comprehend complex systems by modeling them on the basis of high-dimensional data gained from multi-omics technologies. The interdisciplinary challenge resides in applying the right mathematical system to a well-selected and robust set of clinical and experimental data, and to iteratively refine the algorithms by validation experiments. Ideally, this approach will yield computational systems that support decision-making in disease prevention and diagnosis, as well as allow to extract crucial data from large high-throughput datasets, with the ultimate goal to reduce the socio-economic burden posed by pneumonia and other major diseases.