Urinary tract infections are the main reason for consultation in the pediatric emergency department worldwide, so it deserves to be analyzed with artificial intelligence techniques to discover patterns based on medical and laboratory information. Cluster analysis is an unsupervised machine learning technique that allows the identification of groups of patients with similar characteristics. In this work we analyzed information from patients whose anonymized information was extracted from a computer system, all of them are patients suffering from urinary tract infections. Multiple Correspondence Analysis was initially applied and then K-means and DBSCAN algorithms were used separately. The silhouette value of each group identified with the two algorithms was obtained. Patients were differentiated according to the prevalence percentages of sensitivity/resistance to certain antibiotics and t...

Urinary tract infections are the main reason for consultation in the pediatric emergency department worldwide, so it deserves to be analyzed with artificial intelligence techniques to discover patterns based on medical and laboratory information. Cluster analysis is an unsupervised machine learning technique that allows the identification of groups of patients with similar characteristics. In this work we analyzed information from patients whose anonymized information was extracted from a computer system, all of them are patients suffering from urinary tract infections. Multiple Correspondence Analysis was initially applied and then K-means and DBSCAN algorithms were used separately. The silhouette value of each group identified with the two algorithms was obtained. Patients were differentiated according to the prevalence percentages of sensitivity/resistance to certain antibiotics and t...