The spatial characterization of tumor regions depending on the biological heterogeneity of the tissues is an interesting issue for the monitoring and treatment of the illnesses along their evolution. Tissue signatures are unique patterns that can be found in medical images and that are relative to histopathological, mollecular and biological characteristics of the tumor. Therefore, tissue signatures are not morfological segmentations, but sub-compartments or habitats defined inside the morfological regions of the tumor. Manual procedures to delineate complex tissue signatures volumes are not plausible, so the signatures definition must be done using multiparametrical non-supervised methods. In previous results, we have demonstered that functional biomarkers of parametrical tissue signatures extracted from pre-quirurgical images are able to predict global survivance. In this project, we will check the temporal evolution of multiparametrical tissue signatures over the set of assisted patients in the Comunidad Valenciana from 2012 until today. This study will allow to exhaustively characterize the multiparametrical tissue signatures pattern of Glioblastoma, and its temporal evolution.