Bacterial resistance to antibiotics poses a major threat to global health, intensified by the improper use of antibiotics. The lack of effective and rapid diagnostic tools exacerbates the problem, resulting in less efficient treatments and increased spread of bacterial resistance. Therefore, there is an urgent need to develop models that can assist in more precise and restricted antibiotic prescriptions. Thus, the overall objective of this work is to aid in the construction of binary classification models aimed at determining whether a patient is infected by a given bacterium based on clinical and laboratory data. For this purpose, we used data from the Hospital das Clınicas at UFMG. Throughout this work, we will carry out a process of organizing and cleaning the data to extract both patient features and bacterial targets identified through bacterial culture results. To achieve this, we held weekly meetings with doctors from the Hospital das Clınicas, specialists in the hospital’s infectious disease area. This work enabled us to build primary databases that served as input and output for a Gradient Boosting-based model, trained using ROC (Receiver Operating Characteristic) AUC (Area Under Curve) as the optimization metric. Throughout this work, we underwent an extensive data cleaning and aggregation process to ensure that the data could serve as a foundation for this and other Artificial Intelligence (AI) studies in healthcare. Given the complexity of the problem, our models achieved reasonable results.