Using the databases that contain information from patients diagnosed with pulmonary tuberculosis, from the Santa Clara hospital, a comparison was made of 3 clustering algorithms based on neural networks such as: self-organized maps, ART networks and Fuzzy Art networks, to conclude Which of these presents the best grouping index, supporting the diagnosis of the disease. After this, the visualization of the behavior of the neurons that are activated with respect to the condition of discharge from the hospital is carried out, allowing the health professional to reduce the observation time of all his patients and focusing on those at higher risk
