Medical data stands out as one of the most valuable sources of information in contemporary data mining research. These datasets encapsulate diverse information, from patient records to current situations. Simultaneously, this wealth of information poses a challenge for data mining due to its diversity and the requirement for context to discern which data holds value and what conclusions truly contribute to generating new knowledge. In this work, we leverage a database of medical records from patients’ visits to several hospitals across several years, enhancing it with hospital information and a disease ontology. This allows us to identify a medical diagnosis at various levels of semantic depth. Based on this information, we propose information mining centered around extracting association rules at progressive specific levels concerning diagnoses. Subsequently, we present the initial results of this study for different sets of diseases and suggest the most relevant steps for development based on these outcomes.