Krause, Rolf Gonzalez, Santiago Fernandez Pizzagalli, Diego Ulisse 2020-02-19 The immune system has a critical role in diseases of primary importance such as infections and cancer. Hence, it represents a target for novel therapeutic strategies. However, the immune system relies on a complex network of cell-to-cell interactions which remains largely unknown, or difficult to be interpreted. The combination of experimental data with computational methods is of paramount importance to analyze these interactions. Indeed, recently established 2-photon intravital microscopes (2P-IVM), can capture videos of cells while interacting in organs of living animals. These interactions are often associated with specific movement patterns. Hence, computer vision methods have the potential to extract knowledge from these videos by analyzing the movement of cells. Unfortunately, common analysis methods poorly apply to 2P-IVM videos capturing the cells of the immune system. This is mainly due to the complex appearance and biomechanical properties of these cells, as well as challenges introduced by in vivo imaging. Additionally, a lack of publicly available 2P-IVM datasets hampers the development of novel analysis methods along with data-driven studies of the immune system. Finally, common measures of cell motility, poorly describe the dynamic behavior of immune cells. In this thesis, we address these limitations by • Making available the first database of 2P-IVM videos and tracks of immune cells. • Modeling as graph the content of 2P-IVM videos, from pixels to biological processes. • Developing, refining, and applying a variety of computational methods to extract knowledge from this graph. • Shifting the analysis of cell motility towards the recognition of cell actions, which does not necessarily require cell tracking. This combination of microscopy data, graph-based methods, and action-based models allowed us to quantify the complex movement patterns of neutrophils, revealing different phases of the immune response to influenza vaccination. application/pdf https://localhost:5000/ark:/12658/srd1319237 https://susi.usi.ch/global/documents/319237 https://susi.usi.ch/documents/319237/files/2020INFO018.pdf eng info:eu-repo/semantics/altIdentifier/urn/urn:nbn:ch:rero-006-118977 info:eu-repo/semantics/altIdentifier/ark/12658/srd1319237 info:eu-repo/semantics/openAccess License undefined Systems biology Immunology Cell behavior Graph-based methods Action recognition Microscopy info:eu-repo/classification/udc/004 A hierarchy of graph-based methods to study the behavior of immune cells in vivo http://purl.org/coar/resource_type/c_db06