Colorectal cancer is one of the diseases with the highest rate of death worldwide, so it is necessary to improve its diagnosis and treatment. In this sense, new artificial intelligence techniques have emerged as a new way to understand the intrinsic mechanisms of the disease. In the present study, data analysis and machine learning techniques have been applied with the aim of identifying clinical and biological patterns that allow for optimizing patient stratification and improving the treatments administered. Through data cleaning and analysis, distinct patient profiles have been defined, clearly differentiating those with a poorer prognosis from those with a better outlook, thanks to clustering algorithms and the interpretation of the resulting clusters supported by the LIME technique. In addition, a supervised analysis has been performed, also supported by LIME to improve interpretability, which revealed the most influential variables in predicting key clinical events such as survival, tumor progression, and the pathological response of the tumor to treatment. These results highlight the importance of monitoring molecular markers and hematological parameters, as well as ensuring the proper management of treatment timelines, laying the foundation for improving clinical decision-making and advancing toward more personalized medicine.