Redesigning COVID-19 Care With Network Medicine and Machine Learning

Emerging evidence regarding COVID-19 highlights the role of individual resistance and immune function in both susceptibility to infection and severity of disease. Multiple factors influence the response of the human host on exposure to viral pathogens. Influencing an individual’s susceptibility to infection are such factors as nutritional status, physical and psychosocial stressors, obesity, protein-calorie malnutrition, emotional resilience, single-nucleotide polymorphisms, environmental toxins including air pollution and firsthand and secondhand tobacco smoke, sleep habits, sedentary lifestyle, drug-induced nutritional deficiencies and drug-induced immunomodulatory effects, and availability of nutrient-dense food and empty calories. This review examines the network of interacting cofactors that influence the host-pathogen relationship, which in turn determines one’s susceptibility to viral infections like COVID-19. It then evaluates the role of machine learning, including predictive analytics and random forest modeling, to help clinicians assess patients’ risk for development of active infection and to devise a comprehensive approach to prevention and treatment.