Application of neural nets for the identification of otorhinolaryngological diseases

Research devoted to the discovery and identification of diseases in an automated way occur more frequently today. For such, the use of artificial intelligence is conditioned to the learning of the machine, allowing the computer to identify characteristics and learn from new cases, generating new models and identifying in a precise way what it was trained to do. The use of artificial intelligence is not a recent theme but, given the accessibility made available by some platforms that get around the complexity of the algorithm, the theme gained momentum and made possible the use of this technology. This research made use of this technology applied to the identification of three diseases in the field of otolaryngology. The objective was training the computer with the symptoms of these diseases and, then, processing the realization of tests, informing symptoms to the computer and obtaining the answer of the disease it interpreted. After his technique has been applied and some symptoms has been run by it, the computer got the disease right in 90,72% of cases. We intend to expand this research to create models that contemplate more diseases and also make available the interface that shows to the doctor which is the possible disease for that patient.