González Aranda, Ana2024-06-262024-06-262024-06-20https://hdl.handle.net/10115/35104Trabajo Fin de Grado leído en la Universidad Rey Juan Carlos en el curso académico 2023/2024. Directores/as: Pablo Delgado Bonet, Ángel Torrado CarvajalCanine gliomas are a group of neurological neoplasms that account for 2-5% of canine cancers. This incidence is relatively low in comparison with other canine neoplasms but they are significant due to their aggressiveness and difficulty of treatment. In recent years, due to the advances in imaging techniques and the rise of techniques based on machine learning, there has been a change in the way of understanding medical and veterinary diagnosis. The purpose of this project has been to develop a classification tool capable of predicting the response of several canine patients to the proposed treatment, as well as another regression tool to predict the survival time to such treatment and, finally, an evaluation method that indicates how each tumour has responded to treatment at volumetric level. In addition, each of the algorithms and results obtained throughout the project have been explained. This approach is enhanced by the potential of imaging biomarkers as a diagnostic tool, as it can provide critical insights into tumor characteristics and treatment responses, thereby improving the precision and effectiveness of therapeutic strategies and enabling better prognostic evaluations in veterinary oncology.engCreative Commons Atribución-CompartirIgual 4.0 Internacionalhttps://creativecommons.org/licenses/by-sa/4.0/legalcodeRadiomicsPredictive biomarkersCanine GliomaRandom ForestOncolytic virusesICOCAV-15MRISlicer3Dinfo:eu-repo/semantics/studentThesisinfo:eu-repo/semantics/openAccess