AUTHOR=Cohen Irun R. , Efroni Sol TITLE=The Immune System Computes the State of the Body: Crowd Wisdom, Machine Learning, and Immune Cell Reference Repertoires Help Manage Inflammation JOURNAL=Frontiers in Immunology VOLUME=Volume 10 - 2019 YEAR=2019 URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2019.00010 DOI=10.3389/fimmu.2019.00010 ISSN=1664-3224 ABSTRACT=Here, we discuss features of the mammalian immune system that enable it to orchestrate variable inflammatory responses that maintain and heal the body while protecting it from pathogenic infectious agents and tumors. The paper makes four points: 1) The immune system classifies healthy and pathologic states of the body deploying individual lymphocytes as cellular computers; immune cells transform input signals from the body into an output of specific immune reactions. 2) Groups of individually activated immune-system cells co-react in lymphoid organs to make collective decisions through a type of self-organizing swarm intelligence or crowd wisdom. 3) Collective choices by swarms of immune cells, like those of schools of fish, are modified by relatively small numbers of individual regulators responding to shifting conditions – inflammatory responses are shaped dynamically on the run. 4) Immune system decisions are primed by formative experience with training sets of self-antigens encountered during primary development; these initially trained T cell and B cell repertoires then guide later immune responses to test sets of antigens encountered during life in the world. Thus, the immune system uses a kind of experience-based machine learning strategy analogous to that deployed by machine-learning algorithms. This view of the immune system bears clinical implications for monitoring wellness and for treating autoimmune disease, cancer, and allograft reactions.