Abstract of Thesis presented at COPPE/UFRJ as a partial fulfillment of the requirements for the degree of Master of Science (M.Sc.)

Neural Networks in Determination of Chronic Arterial Hypertension and Periodontal Disease Using Salivary Metabolites as Indicators

Adriano Gomes Sabino de Araujó

September/2014

Advisors:

Luiz Pereira Calôba Carla Martins de Oliveira

Department:

Eletrical Engineering

      Artificial Neural Networks are computational techniques that are based on the structure and functioning of the brain and therefore try to take advantage of features such as his ability to learn and recognize patterns quickly and efficiently. These techniques have been widely used to perform complex tasks such as object detection in images and text, voice recognition and various types of predictions. The disease forecasting is one application of neural networks that are being constantly discussed in the literature. Based on this, the objective of this work is to use such techniques to try to predict the Hypertension and Chronic Periodontal Disease from salivary metabolites of women in the postpartum period identified by Nuclear Magnetic Resonance (NMR). The prediction results were reasonable for Chronic Hypertension and good for Periodontal Disease, indicating the existence of an association between these diseases and salivary metabolites.